Proceedings of 9 th Windsor Conference: Making Comfort Relevant Cumberland Lodge, Windsor, UK, 7-10 April 2016. Network for Comfort and Energy Use in Buildings, http://nceub.org.uk Thermal Comfort with Convective and Radiant Cooling Systems Risto Kosonen 1, Panu Mustakallio 2, Zhecho Bolashikov 3 and Arsen Melikov 3 1 Aalto University, Finland, risto.kosonen@aalto.fi 2 Halton Oy, Finland 3 International Centre for Indoor Environment and Energy, Technical University of Denmark Abstract The thermal environment in a two person office room obtained with chilled beam (CB), chilled beam with radiant panel (CBR), chilled ceiling with ceiling installed mixing ventilation (CCMV) and four desk partition mounted local radiant cooling panels with mixing ventilation (MVRC) was compared with laboratory measurements. CB provided convective cooling while the remaining three systems (CBR, CCMV and MVRC) provided combined radiant and convective cooling. Solar radiation, office equipment, lighting and occupants were simulated to obtain different heat load conditions; 38 W/m 2 and 64 W/m 2. Air temperature, operative temperature, radiant asymmetry, air velocity and turbulent intensity were measured and draught rate levels calculated. The results revealed that the differences in thermal conditions achieved with the four systems were not significant. CB and CBR provided slightly higher velocity level in the occupied zone. Keywords: physical measurements, radiant cooling, convective cooling, thermal comfort 1 Introduction Present standards (ISO 7730, 2005 and EN 15251, 2007) recommend maximum values for the indoor climate parameters for both winter and summer conditions in order to achieve thermally comfortable environment for occupants. However, in rooms with high cooling demand (> 50 W/m 2 -floor area) it becomes challenging to achieve the targeted indoor climate without sacrificing occupants thermal comfort due to the increased convective flows. Therefore cooling systems based on convective, radiant or combined heat exchange are used. The differences between radiant and convective systems have been discussed, and in the literature and by system manufacturers, often radiant systems have been recommended for being able to provide favourable difference in the operative temperature for energy efficiency and thermal comfort, and more acceptable draught rate levels when comparing to convective systems. This study was performed to compare the performance of four systems based on radiant and convective cooling, namely chilled beam (CB), chilled beam with radiant panel (CBR), chilled ceiling with ceiling installed mixing ventilation (CCMV) and four desk partition mounted local radiant cooling panels with ceiling installed mixing ventilation (MVRC). The CB and CCMV systems are often used in modern office buildings. The study comprised comprehensive physical measurements. This paper compares the physical environment obtained with the systems in two person office room. The results of the Windsor Conference 2016 - Making Comfort Relevant - Proceedings 322 of 1332
human subject experiments performed in the case of office room are presented in a separate paper (Duszyk et al., 2011). 2 Method Measurements were performed in a climate chamber (4.12 x 4.20 x 2.89 m, L x W x H) under steady state conditions at 26 C design room air temperature. Office room with two occupants were simulated. Two cooling conditions were simulated: design (maximum) heat load 64 W/m 2 and usual heat load 38 W/m 2. Heat load from occupants, computers, four lighting units and solar radiation was simulated. The heat load for the studied conditions is specified in Table 1. Surface temperature of simulated windows (water panels) and floor surface temperature (electrically heated foils below part of floor covering near the windows) was controlled to distribute the solar heat gain. Table 1. Heat balance in measured cases. Heat balance of test for Office room In cooling conditions with Maximum heat loads Usual heat loads Occupants (about 78 W/occupant) 2 persons 2 persons 156 W 156 W 9 W/m² 9 W/m² Computers (about 65 W/computer) 2 computers 2 computers 130 W 130 W 8 W/m² 8 W/m² Lighting 160 W 160 W 9 W/m² 9 W/m² Solar load - window surface temperature 34 degc 30 degc with 6.3 m2 window and 26 degc room ~ 404 W 202 W Solar load - direct solar load on the floor 250 W 0 W Total solar load 38 W 12 W Total heat loads 1100 W 648 W 64 W/m² 38 W/m² Supply air flow rate 26 l/s 26 l/s Supply air temperature 16 degc 16 degc Supply air cooling power in 26 degc room 312 W 312 W 18 W/m² 18 W/m² Cooling power demand from water 788 W 336 W 46 W/m² 20 W/m² The positioning of the heat load in the room is shown in Figure 1. Air temperature, operative temperature, mean velocity and turbulent intensity were measured at 8 heights (0.05 m, 0.1 m, 0.3 m, 0.6 m, 1.1 m, 1.7 m, 2.0 m, 2.4 m from floor), at 25 locations in the room (Figure 1). At the locations of the desks, the measurements were performed at heights 1.1, 1.7, 2.0 and 2.4 m. Surface temperature (walls, floor, ceiling, window) and radiant temperature asymmetry was measured. Draught rate index (ISO 7730, 2005) was calculated based on the measured parameters. Radiant asymmetry was measured at 1.1 m height at location 13 (Fig. 1) in CCMV, CB and CBR cases. In MVRC cases, radiant asymmetry was measured at the workstation at 1.1 m height between both occupants and radiant panels, and at 0.6 m and 1.1 m heights between both occupants and side walls. Air temperature and operative temperature sensors were of a thermistor type with accuracy of ±0.2 C. Air temperature was measured with radiation shielded sensors. Velocity sensors were of an omnidirectional hot-sphere type with accuracy Windsor Conference 2016 - Making Comfort Relevant - Proceedings 323 of 1332
of ±0.2 m/s or ±1% of the reading 0.05-0.5 m/s. Multi-channel wireless low-velocity thermal anemometer with eight velocity sensors was used. All measurement sensors were calibrated prior to the measurements and measurement results were 5 minutes average readings. Figure 1. Top view of the test room with measurement pole locations. The operating principle of the four cooling systems is schematically shown in Figure 2. In the case of CCMV cooling panels were integrated into the false ceiling tiles. The radiant ceiling covered 77% of the total ceiling surface. The top surface of the tiles was not insulated. Supplied air was distributed with two linear diffusers both with two slots size 472 x 20 mm, L x W each (Figure 2). Supply air temperature in all cases was 16 C and water inlet temperature 15 C with return water 2-3 C warmer. The same chilled beam (coil length 2100 mm) was used in the cases CB and CBR was used (Figure 2). In the case CBR the surface area of the radiant panels was 3.6 m². The chilled beam was removed from ceiling when chilled ceiling cases were measured. Personal radiant panels were installed at the desks as shown in Figure 3. In this case with design heat loads, supplied air volume flow was increased to 44 l/s to compensate the missing cooling power from panel radiators. Windsor Conference 2016 - Making Comfort Relevant - Proceedings 324 of 1332
Figure 2. Operating principle of the four cooling systems: A) CCMV, B) CB, C) CBR and D) MVRC 3 Results Summary of measurements results, including average, minimum, maximum and standard deviation of values, has been presented in Table 2 for overview of the thermal conditions. Thermal conditions with all studied systems were very similar and similar behavior of the air distribution can be seen in smoke visualizations for all cases with supply air jets turning towards the wall opposite to simulated window. Average room air temperature and operative temperature were rather similar with all systems; only small difference, on an average 0.2 C, between them can be seen in design conditions. There was significant horizontal operative temperature difference between window side and door side of the room (in design conditions on an average 1.5 C and in usual conditions 1.0 C). The maximum horizontal temperature difference in design conditions was about 2.0 C. It was similar with all cooling systems and was caused by the one-sided locations of the heat loads. Only in MVRC cases this difference was a bit smaller and that was also measured in lower equivalent temperatures of the thermal manikins. Due to the horizontal temperature difference, the operative temperature level near the window was about 0.4-1.8 C higher than room design temperature (in the middle) in all cases. Vertical temperature difference in the room in all cases was very small (-0.1-0.7 C). This difference was a bit smaller with radiant systems CCMV and CBR due to the bigger view factor towards floor when comparing to the MVRC case. In the design cooling case the difference can be seen most clearly. Windsor Conference 2016 - Making Comfort Relevant - Proceedings 325 of 1332
Table 2 Summaries of measurement results in design (64 W/m 2 ) and usual (38 W/m 2 ) conditions. OFFICE ROOM IN DESIGN CONDITIONS OFFICE ROOM IN USUAL CONDITIONS Measurement results in occupied zone at heights 0.1, 0.6, 1.1 and 1.7 m CCMV CB CBR MVRC CCMV CB CBR MVRC Average air temperature [ C] 26.1 25.8 26.1 25.9 26.0 25.8 25.9 25.8 Min. air temperature [ C] 25.3 24.9 25.3 25.0 25.5 25.0 25.2 24.2 Max. air temperature [ C] 28.0 26.7 27.6 27.2 26.8 26.4 26.7 26.7 Std. dev. of air temperature [ C] 0.6 0.5 0.6 0.4 0.3 0.3 0.3 0.4 Average operative temperature [ C] 26.3 26.1 26.3 26.1 26.1 25.9 26.1 25.9 Min. operative temperature [ C] 25.3 25.1 25.4 25.4 25.5 25.1 25.2 24.5 Max. operative temperature [ C] 28.2 27.2 27.8 27.5 27.1 26.6 26.9 26.8 Std. dev. of operative temperature [ C] 0.7 0.6 0.6 0.5 0.3 0.4 0.4 0.4 Average operative - air temperature [ C] 0.1 0.3 0.2 0.1 0.1 0.1 0.0 0.1 Min. operative - air temperature [ C] -0.2-0.1 0.0 0.0-0.1-0.1-0.3-0.1 Max. operative - air temperature [ C] 0.6 0.7 0.5 0.6 0.5 0.4 0.3 0.3 Std. dev. of operative-air temperature [ C 0.2 0.2 0.1 0.1 0.1 0.1 0.1 0.1 At height 1.1 m: Avrg. air temperature of window side [ C] 26.8 26.4 26.9 26.8 26.4 26.2 26.4 26.5 Avrg. air temperature of door side [ C] 25.7 25.4 25.7 25.9 25.7 25.6 25.7 25.9 Avrg. horizontal air temp. diff. [ C] 1.1 1.0 1.2 1.0 0.7 0.7 0.7 0.6 Avrg. oper. temp. of window side [ C] 27.4 27.1 27.4 27.3 26.8 26.6 26.7 26.7 Avrg. oper. temperature of door side [ C] 25.8 25.7 25.9 25.9 25.9 25.7 25.8 26.0 Avrg. horizontal oper. temp. diff. [ C] 1.6 1.4 1.5 1.3 0.8 0.9 0.9 0.8 Avrg. oper. - air temp. of window side [ C] 0.6 0.7 0.5 0.4 0.4 0.3 0.3 0.2 Avrg. oper. - air temp. of door side [ C] 0.1 0.2 0.2 0.1 0.2 0.1 0.0 0.0 At heights 0.1 m - 1.7 m: Avrg. vertical air temperature diff. [ C] 0.0 0.3 0.2 0.6 0.3 0.4 0.2 0.8 Avrg. vertical oper. temperature diff. [ C] -0.1 0.5 0.2 0.6 0.3 0.5 0.5 0.7 Max. radiant asymmetry (window-door) [ C] 5.0 4.0 4.2 3.3 2.3 3.2 2.5 2.3 Max. radiant asymmetry (side-side wall, except in MVRC radiator-side wall) [ C] 0.3 0.6 1.5-3.4 0.7 0.7 0.8-1.7 Max. radiant asymmetry (floor-ceiling, except in MVRC radiator-manikin) [ C] 4.1 0.8 1.7-6.5 3.0-0.8 0.3-4.3 Average air velocity [m/s] 0.13 0.13 0.12 0.10 0.11 0.12 0.11 0.06 Average of 3 highest velocities [m/s] 0.23 0.27 0.24 0.20 0.21 0.26 0.25 0.13 Highest velocity [m/s] 0.24 0.29 0.25 0.21 0.23 0.27 0.26 0.14 Std. dev. of air velocity [m/s] 0.04 0.05 0.05 0.04 0.04 0.05 0.05 0.03 Average turbulence intensity [%] 39 45 45 47 40 42 48 46 Average of 3 highest turb. intensities [%] 76 74 77 78 71 72 84 100 Std. dev. of turbulence intensity [%] 14 13 12 15 12 15 16 58 Average draught rate [%] 7.9 9.5 8.1 6.1 5.7 7.8 6.9 2.0 Average of 3 highest draught rates [%] 14.7 19.8 18.0 14.1 12.3 18.2 16.6 7.2 Highest draught rate [%] 16.3 20.8 19.5 14.9 13.0 18.4 17.4 8.3 Std. dev. of draught rate [%] 3.3 4.6 4.2 3.7 2.8 4.6 4.2 2.1 The radiant asymmetry was in all cases between window and door wall few degrees Celsius, so quite significant due to the solar heat load and supply air jet cooling the door wall when turning towards it. Also in CCMV case radiant asymmetry between floor and ceiling was at the same level due to the chilled ceiling surface. Radiant asymmetry in MVRC case was measured at different locations than in other cases. The radiant asymmetry in MVRC case was few degrees between all directions it was measured, and it was highest between radiant panel and thermal manikin (6.5 C). Average room air velocities were quite similar with all systems. With design heat loads in CCMV, CB and CBR cases relatively high velocities can be found (0.24-0.34 m/s). When usual heat loads are used, the velocity levels were mostly lower in all cases except in MR case where highest velocities were bigger with CB and CBR systems. This difference with CB and CBR in MR case was however smoothed away when average of three highest velocities was compared. Windsor Conference 2016 - Making Comfort Relevant - Proceedings 326 of 1332
There was quite consistent 0.05 m/s difference in highest room air velocity levels between radiant CCMV system and purely convective CB system. The highest velocities of MVRC were lower in both design and usual heat load cases when compared with other systems. The average turbulence intensity was rather similar with all systems, between 40% and 50%. Also average of three highest turbulence intensities were at roughly same level. Very high turbulence intensity readings in MVRC case with usual heat loads were caused by very low air velocity level (average 0.06 m/s). The average draught rate difference in measurement pole readings was small, 1-2% higher in purely convective CB cases and the highest readings were about 5% higher in CB cases than in mostly radiant CCMV cases. The effect of using radiant panels integrated chilled beam can be seen slightly in the draught rate results, in CBR cases of OR, but not in MR cases. With usual heat loads, draught rates got smaller for all systems. This was most pronounced in the CCMV and MVRC cases. Draught rate levels in MVRC cases were lower than with other systems. 4 Conclusions The results revealed that the differences in thermal conditions between the measured radiant and convective systems were not big. Also with radiant cooling systems, the convective heat transfer was significant and convective flows in the room were similar. The air temperature and operative temperature were near each other and very similar in all studied cases with about 0.2 C difference between operative and room air temperature. This is differs from some design guides of radiant cooling systems where even 2-3 C lower operative temperature is used. Based on this research this might lead to too high room air temperatures in cooling design conditions. There was significant horizontal operative temperature difference between window side and door side of the room (maximum difference about 2.0 C) with all systems. The room air velocities and draught rates were slightly higher in the CB and CBR cases (about 0.05 m/s difference in the highest velocities and 5 % in the highest draught rates) when compared with CCMV cases. When comparing thermal conditions with the design and usual heat load levels, with higher heat loads levels, the velocity and draught rate levels got higher in all cases. Acknowledgement The study is supported by Technology Agency of Finland (TEKES) in RYM-SHOK research program and the International Centre for Indoor Environment and Energy, Department of Civil Engineering, Technical University of Denmark. References ISO 7730,2005, Ergonomics of the thermal environment - Analytical determination and interpretation of thermal comfort using calculation of the PMV and PPD indices and local thermal comfort criteria. International Organization for Standardization, Geneva, Switzerland. EN 15251,2007, Indoor Environmental Input Parameters for Design and Assessment of Energy Performance of Buildings Addressing Indoor Air Quality, Thermal Environment, Lighting and Acoustics. European Committee for Standardization, B-1050 Brussels. Duszyk M, Melikov A, Kosonen R, Mustakallio P,2011. Comparison of Temperature and Velocity Field in Rooms with Chilled Beams and Radiant Panel Systems Combined with Mixing Ventilation. Proceedings of Roomvent 2011 conference. Trondheim, Norway. Windsor Conference 2016 - Making Comfort Relevant - Proceedings 327 of 1332