ASHRAE LV-11-C011-2011 Prediction of Air Mixing from High Sidewall Diffusers in Cooling Mode.pdf

1、El Hassan Ridouaneg3g76g86g3g68g3g51g82g86g87g71g82g70g87g82g85g68g79g3g53g72g86g72g68g85g70g75g72g85g3g82g73g3g53g72g86g76g71g72g81g87g76g68g79g3g37g88g76g79g71g76g81g74g3g54g92g86g87g72g80g86g3g68g87g3g87g75g72g3g49g68g87g76g82g81g68g79g3g53g72g81g72g90g68g69g79g72g3g40g81g72g85g74g92g3g47g68g69g8

2、2g85g68g87g82g85g92g15g3g42g82g79g71g72g81g15g3g38g82g17g3Keith Gawlikg3g76g86g3g68g3g54g72g81g76g82g85g3g40g81g74g76g81g72g72g85g3g82g73g3g55g75g72g85g80g68g79g3g54g92g86g87g72g80g86g3g68g87g3g87g75g72g3g49g68g87g76g82g81g68g79g3g53g72g81g72g90g68g69g79g72g3g40g81g72g85g74g92g3g47g68g69g82g85g68g87

3、g82g85g92g15g3g42g82g79g71g72g81g15g3g38g50g17g3 g3 g3Prediction of Air Mixing from High Sidewall Diffusers in Cooling Mode El Hassan Ridouane, PhD Keith Gawlik, PhD Member ASHRAE Member ASHRAE ABSTRACT Computational Fluid Dynamics (CFD) modeling was used to evaluate the performance of high sidewall

4、 air supply in cooling mode. The research focused on the design, placement, and operation of air supply diffusers located high on a sidewall and return grilles located near the floor on the same sidewall. Parameters of the study are the supply velocity, supply temperature, diffuser size and room dim

5、ensions. Thermal loads characteristic of high performance homes were applied at the walls and room temperature was controlled via a thermostat. The results are intended to provide information to guide the selection of high sidewall supply diffusers to provide proper room air mixing and thermal comfo

6、rt for cooling of high performance homes. INTRODUCTION As Building America (BA) homes begin to reach the 40 to 50% performance levels, the downsizing of space conditioning systems will begin to reach the point where the air flow volumes required to meet the small remaining heating and cooling loads

7、will not be large enough to maintain uniform mixing of room air. This project focused on resolving this technical barrier so that uniform temperatures can be maintained within future homes even though the cooling and heating capacities required to heat and cool the homes are significantly smaller th

8、an current homes. In addition, well insulated homes with advanced windows do not require perimeter heating, so centralized, compact duct systems in interior walls can be used. Compact systems are 20-25% less expensive than traditional designs, and offer higher system efficiency through reduced duct

9、leakage and duct length, and by having the ducts in conditioned spaces (Griffiths and Zoeller, 2001). High sidewall diffusers are one of the simplest and most promising approaches of delivering conditioned air in very high performance homes since they are not blocked by furniture and, with proper de

10、sign, can provide good mixing of the conditioned air in the space. This report describes simulated performance of high sidewall diffusers for typical residential applications. The task addressed the evaluation of these systems for cases in which the room is recovering from a temperature setback stat

11、e. This was considered to be a worst case operating condition for the high sidewall diffuser system, even though it may not represent the most common operating mode. However, if the system performs well in terms of providing good air mixing during transient setback recovery, it will likely perform w

12、ell during steady-state, part load and design load operation. A design guide to evaluate the performance of high sidewall air supply for residential applications was developed by Temple (2003). Following this guide, representative room and diffuser geometries, air supply flow rates, and temperatures

13、 were identified and modeled using Fluent CFD software. The model addressed thermal loads that are representative of high performance homes. A fixed load density of 10 Btu/h.ft2(31.52 W/m2) was considered in this study. This density was multiplied by the floor area and the outcome was distributed be

14、tween the ceiling and at the exterior wall opposite to the diffuser jet. These loads were distributed based on R-20 wall and R-30 ceiling. A 4 in. (0.1 m) by 4 in. (0.1 m) thermostat with a dead-band between 70F (294.3 K) and 72F (295.4 K) was built into the model to control the room temperature. Th

15、e thermostat was mounted on the wall parallel to the diffuser jet at 4 ft. (1.22 m) from the floor (see Figure 1). g3LV-11-C011 2011 ASHRAE 872011. American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions, Volume 117, Part 1.

16、For personal use only. Additional reproduction, distribution, or transmission in either print or digital form is not permitted without ASHRAES prior written permission.NUMERICAL MODEL The analysis considered two different rooms: Room A is 12 ft (3.66 m) long 11 ft (3.35 m) wide 9 ft (2.74 m) high, s

17、upplied by an 8 in. (0.2 m) 4 in. (0.1 m) diffuser and Room B is 12 ft (3.66 m) long 9 ft (2.74 m) wide 9 ft (2.74 m) high, supplied by a 10 in. (0.25 m) 6 in. (0.15 m) diffuser. The diffusers were based on available products and used the geometry and open area of the J #23#23#23Figure 1 Sketch of t

18、he computational domain (not to scale) showing the locations of supply, return, and thermostat in the room. Thermal loads are applied only to the ceiling and to the right wall Symmetry plane Supply L W/2 H Thermal load Thermal load Thermostat Return 88 ASHRAE TransactionsCOMFORT CRITERIA The air dif

19、fusion performance index (ADPI) was calculated at the end of the simulation when one air change was supplied to the room. ADPI is defined in a zone between 0.1 m and 1.7 m from the floor (ASHRAE 1992) and is a simple comfort criterion based on local temperature, average temperature in the whole room

20、, and local air speed. ADPI uses effective draft temperature, ;#23#23#23)30(07.0)( =xavgVTT ;#23#23#23;#23#23#23;#23#23#23;#23#23#23;#23#23#23;#23#23#23;#23#23#23;#23#23#23(1);#23#23#23;#23#23#23;#23#23#23where T is the local air temperature F, Tavgis the average temperature in the occupied zone F,

21、and Vxthe local air speed fpm. A high percentage of occupants are comfortable in indoor environments when the effective draft temperature is between -3F (-1.5C) and 2F (1C) and local air speed below 70 fpm (0.35 m/s) for maximum comfort (ASHRAE 2009). In the zone of interest, effective draft tempera

22、ture and air speed were determined for each cell. The number of cells meeting these requirements was compared to the total number of cells to determine the ADPI for the simulation. Acceptable comfort level exists in the room when ADPI is in the range between 70 and 90%. Good comfort level is guarant

23、eed when the ADPI is 90% or higher. RESULTS AND DISCUSSION The results will be presented first for Room A and then for Room B by the temperature distribution and comfort level for different operating conditions. The time history of the thermostat temperature in Room A supplied by TSof 55F (285.9 K)

24、is presented in Figure 2 for two different supply velocities. The corresponding velocity variations at the diffuser are plotted in Figure 3. The time at which one air change was supplied to the room is marked on Figure 2. The temperature at the thermostat decreased quickly with the supply velocity o

25、f 591 fpm (3 m/s) and entered a quasi-steady state mode characterized by oscillation cycles around the thermostat deadband. A decrease in supply velocity to 394 fpm (2 m/s) delayed the room cooling and resulted in longer cooling/heating cycles. We are interested in the flow patterns at the time when

26、 one air change was supplied to the room. This is presented in Figure 4 in terms of the distribution of temperature corresponding to VS= 788 fpm (4 m/s) and TS= 55 F (285.9 K). The contour plots are on the vertical mid-plane of symmetry at 25%, 50%, 75%, and 100% of the time for one air change. The

27、isotherms showed a good mixing between the supply air and the room air. The jet was attached to the ceiling and mixed with the room air after hitting the opposite wall. Low supply velocities resulted in sinking jets, which may cause occupant discomfort. ;#23#23#23;#23#23#23Figure 2 Temporal evolutio

28、n of the thermostat temperature in Room A when supplied by TS= 55F (285.9 K). One air change g670;#23#23#23deadband g670;#23#23#23 2011 ASHRAE 89;#23#23#23;#23#23#23Figure 3 Temporal evolution of the supply velocity at the register in Room A when supplied by TS= 55F (285.9 K). ;#23#23#23;#23#23#23 ;

29、#23#23#23;#23#23#23;#23#23#23;#23#23#23;#23#23#23 ;#23#23#23;#23#23#23;#23#23#23Figure 4 Temperature distribution at the symmetry plane of Room A when VS= 788 fpm (4 m/s) and TS= 55F (285.9 K). A different view of the flow structure is illustrated in Figure 5 at selected cross sections along the roo

30、m. This was at an intermediate supply velocity VS= 591 fpm (3 m/s). The left image was for TS= 55F (285.9 K) and the right image was for TS= 65F (291.5 K), both at one air change. The temperature fields showed well mixed air within the room with some hot regions near the ceiling and at the wall oppo

31、site to the diffuser throw. These hot spots, which were a result of thermal loads applied at these locations, were more intense at higher supply temperature of 65F (291.5 K). Figure 6 shows the plots of acceptable draft temperature, between -3 and 2F, on the top row and the plots of draft temperatur

32、e at full scale on the bottom row. Three supply velocities of 394 fpm (2 m/s), 591 fpm (3 m/s), and 788 fpm (4 m/s) were tested and the ADPI was evaluated at one air change. The red line limits the occupied zone. Regions of acceptable draft temperature are larger at low velocity and decrease as the

33、velocity increases. As a result, the velocities of 394 fpm (2 m/s) and 591 fpm (3 m/s) provided good occupant comfort with ADPIs of 97% and 92%, respectively. T (F) ;#23#23#2325% of one air change 50% of one air change 75% of one air change 100% of one air change 90 ASHRAE Transactions;#23#23#23;#23

34、#23#23;#23#23#23;#23#23#23;#23#23#23;#23#23#23;#23#23#23 ;#23#23#23;#23#23#23;#23#23#23;#23#23#23Figure 5 Temperature distribution in Room A at one air change when supplied by VS= 591 fpm (3 m/s): (a) TS= 55F (285.9 K) and (b) TS= 65F (291.5 K). ;#23#23#23;#23#23#23;#23#23#23;#23#23#23 ;#23#23#23;#2

35、3#23#23;#23#23#23;#23#23#23;#23#23#23;#23#23#23;#23#23#23;#23#23#23 ;#23#23#23;#23#23#23;#23#23#23Figure 6 Distribution of draft temperature in Room A when TS= 55F (285.9 K): (a) VS= 394 fpm (2 m/s), (b) VS= 591 fpm (3 m/s), and (c) VS= 788 fpm (4 m/s).Top row shows acceptable draft temperature, bet

36、ween -3 and 2F, and the bottom row shows the draft temperature at full scale. (a): TS= 55F (285.9 K) (b): TS= 65F (291.5 K) T (F) ADPI = 92% ADPI = 69% ADPI = 97% (a)(b) (c) 2011 ASHRAE 91The simulation results for Room B are discussed in the remainder of this section. This room is relatively smalle

37、r than Room A and is supplied by a larger diffuser. At a fixed supply velocity, one air change will be reached in Room B much faster than in Room A. Illustrated in Figure 7 is the variation with time of the thermostat temperature of Room B. Two supply air temperatures of 55F (285.9 K) and 65F (291.5

38、 K) sharing a supply velocity of 788 fpm (4 m/s) were considered. Both curves started at a high value of about 80F (299.8 K) and decreased monotonically with time to reach a minimum of about 70F (294.3 K) and then entered an oscillation mode as a result of thermostat control. The large temperature v

39、alues at the start of the simulation were attributed to the initial condition of uniform room temperature of 80F (299.8 K). With a supply of 55F (285.9 K), the room cooled quickly and stayed within the comfort zone set by the thermostat. However, with 65F (291.5 K) air provided at the diffuser, the

40、cooling took longer and for short periods of time, the room temperature went above the comfort zone. ;#23#23#23;#23#23#23Figure 7 Temporal evolution of the thermostat temperature in Room B when supplied by VS= 788 fpm (4 m/s). The flow visualization in Room B is presented in Figure 8 by the distribu

41、tions of temperature and velocity at the symmetry plane. Three velocities of 329 fpm (1.67 m/s), 558 fpm (2.83 m/s), and 788 fpm (4 m/s) supplying 55F (285.9 K) air were simulated. At low velocity of 329 fpm (1.67 m/s), the air jet lost momentum towards the middle of the room and sank into the occup

42、ied zone, which may result in occupants discomfort. A different scenario was observed at 558 fpm (2.83 m/s) and 788 fpm (4 m/s), where the jet had enough momentum to cross the room and circulate along the wall. Better mixing occurred at the higher velocity of 788 fpm (4 m/s) as depicted by the veloc

43、ity vectors in Figure 8(d). Comfort issues are discussed here by measuring the ADPI in the occupied zone of Room B. The outcome is shown in Figure 9 for the supply temperature of 65F (291.5 K) and velocities of 329 fpm (1.67 m/s), 558 fpm (2.83 m/s), and 788 fpm (4 m/s). The plots show cells in the

44、occupied zone with acceptable draft temperature and air speed. The occupied zone, which was defined above, is limited by the red line on each plot. Similar to the results obtained for Room A, ADPI was higher at low velocities and decreased as the velocity increased. Room B presented lower comfort le

45、vels at high velocities when compared to Room A. Despite the good mixing obtained at high velocity, ADPI decreased due to the condition on air speed to be below 70 fpm. ;#23#23#23g670;#23#23#23g670;#23#23#23One air change 1/3 of one air change 92 ASHRAE Transactions;#23#23#23;#23#23#23;#23#23#23;#23

46、#23#23 ;#23#23#23 ;#23#23#23;#23#23#23;#23#23#23;#23#23#23;#23#23#23 ;#23#23#23 ;#23#23#23;#23#23#23;#23#23#23Figure 8 Temperature distribution (a-c) and velocity vectors (d) at the symmetry plane of Room B: (a) VS= 329 fpm (1.67 m/s), (b) VS= 558 fpm (2.83 m/s), and (c-d) VS= 788 fpm (4 m/s). The s

47、upply temperature is 55F (285.9 K). ;#23#23#23 ;#23#23#23 ;#23#23#23;#23#23#23;#23#23#23;#23#23#23Figure 9 Cells with acceptable draft temperature and air speed, in the occupied zone of Room B, when TS= 65F (291.5 K): (a) VS= 329 fpm (1.67 m/s), (b) VS= 558 fpm (2.83 m/s), and (c) VS= 788 fpm (4 m/s

48、). CONCLUSIONS The application of high sidewall supply diffusers in residential buildings was simulated in cooling mode. Two diffuser dimensions, based on available products, were considered and evaluated for a wide range of operating conditions. Thermal ADPI = 97%ADPI = 93% ADPI = 53%(a) (b) (c) ;#

49、23#23#23T (F)(a) (b) (c) (d) 2011 ASHRAE 93loads characteristic of high performance homes were applied at the walls of the room and the airflow was controlled by a thermostat. The three-dimensional numerical model was solved using the finite volume method under the platform of FLUENT software. The simulations ended when one air change is supplied to the room. The effect of thermostatic control and thermal loads on the behavior of these systems is summarized in Figure 10 by the variations of the ADPI as a function of ai