Indoor climate of an unheated apartment and its impact on the heat consumption of adjacent apartments TEET-ANDRUS KÕIV, ANTI HAMBURG, MARTIN THALFELDT, JEVGENI FADEJEV Department of Environmental Engineering Tallinn University of Techlogy Ehitajate tee 5, 19086 Tallinn ESTONIA Teet.Koiv@ttu.ee http://www.ttu.ee Abstract: - During refurbishing apartment buildings thermostatic radiator valves (TRVs) are often installed in front of each radiator to maximize saving. However, inhabitants often do t kw how to use TRVs and in order to motivate them to learn it, heat allocation systems are used. These systems motivate people to lower indoor air temperature in their apartment, but this increases the heat consumption of the surrounding apartments. This article studies how much changing indoor temperature in an apartment causes heating consumption fluctuations in adjacent apartments and also how much the temperature drops in an apartment if the TRVs are shut off. The analysis has been done with building simulations for one t refurbished and two refurbished cases. The results show that an unheated apartment located in the middle of a building may increase the heat consumption of the adjacent apartments as much as 20%. Only in case of a t refurbished building the temperature drops as low as +14.1 C and is lower than +18 C for 3400 hours per year. If an apartment building has been refurbished, the temperature in the apartment analyzed does t drop significantly below +18 C. It was also concluded that the drop in the total heat consumption of an apartment building is significantly lower than the drop in the heating consumption of a single unheated apartment. Key-Words: - Energy efficiency, static heat flows, heat consumption, indoor climate, thermostatic radiator valves 1 Introduction Increasing efficiency of buildings has become one of the big problems that must be solved. All EU member states have been set the goal of 20% reduction of carbon dioxide (CO 2 ) emission by the year 2020. As the residential sector is responsible for a large part of consumption, it is clear that refurbishing dwellings is an effective means of reducing CO 2 emissions. In case of old apartment buildings, space heating is the single most contributor to greenhouse gas emissions [1]. The heating consumption of dwelling space can mainly be reduced by insulating the envelope and installing windows with low heat conductivity. As the heat losses of buildings are reduced, thermostatic radiator valves (TRVs) must be installed. Studies show that TRVs can contribute greatly to reducing heating consumption, however it is also ted that they must be used correctly [2, 3]. The use of thermostats has been studied by Karjalainen and it was concluded that due to different reasons, many people do t kw how to use TRVs [4, 5]. One means of drawing more dwelling residents attention to TRVs is implementing allocation systems that record consumption and divide it between flats [6]. The system motivates inhabitants to learn how to use TRVs, as the residents have some control over their bills. However, besides benefits, the system can cause problems. In a review of ventilation in European dwellings, it was pointed out that ventilation of residential spaces in Nordic countries is often poor [7]. As a result of refurbishing, the air tightness of the building envelope increases and as many buildings have natural ventilation, the air change rate is reduced. By using allocation systems dwelling inhabitants are motivated to close fresh air valves and t use fans, if they have been installed, thereby furthermore reducing ventilation efficiency. A study showed that from 200 apartments, where a heating allocation system had been installed, 68% did t use the installed fans and the most common reason for that was elevated heating costs [8]. By using heat allocation systems the inhabitants are also motivated to reduce the indoor air temperature of their apartments. As the heat conductivity of interior walls and ceilings is significantly higher than of exterior ones, heat flows between apartments ISBN: 978-1-61804-132-6 52
occur and they are difficult to estimate [6]. As a result the heat losses of an apartment can be eventually compensated by the heat emitted by radiators of ather one, thus causing inequality. The results of experiments conducted by Parsons suggested that, by adjusting clothing, thermal comfort indoors can be achieved within the temperature range of +18 +27 C [9]. The lower limit +18 C is significantly lower than the rmally accepted indoor temperature of +21 C. Furthermore, in the literature survey of the influence of different factors on human comfort in indoor environments it was pointed out that in naturally ventilated buildings the acceptable temperature range is even wider than in buildings with mechanical ventilation [10]. The effect of temperature differences between apartments on bills is often underestimated. The aim of this study is to find out how much changing indoor temperature in an apartment causes heating consumption fluctuations in adjacent apartments and also how much the temperature drops in an apartment if the TRVs are shut off. The analysis has been done by simulating the consumption and indoor climate of a simple 3-story apartment building. Besides analyzing the consumption of different apartments, the reduction in the total heat consumption of the building was also studied. 2 Situation and methods An apartment building located in Estonia and built in the 1980s with 3 stories and 18 apartments was used in the study. The heated area of the building is 1195 m 2. The plan of a typical floor of the building with apartment numbers is given in figure 1. All 3 stories with apartments are similar. Figure 1 Plan on the building The analysis focused on apartment 9 and the apartments surrounding it. Initially a situation, in which the heating temperature setpoint of all apartments was +21 C, was simulated and then the TRVs of the apartment were shut off. The indoor air temperatures of apartment 9, the heat consumption of the adjacent apartments and the whole building were analysed. It must also be ted that the basement is unheated. The simulation was done for typical Estonian heating period from September 1 st to May 31 st. 3 different cases were studied: A. Initial t refurbished situation with natural ventilation B. A refurbished building with mechanical exhaust ventilation C. A refurbished building with ventilation heat recovery The description of variants A, B and C is given in table 1 and the heat conductivities of building structure elements is given in table 2. Table 1 Description of building variant A, B and C A B C Ventilation in apartments, air change/flow rate Infiltration, l/(sm 2 ext. surf) Specific heat loss per heated area, W/(K m 2 ) Natural 0.5 1/h Mechanica l exhaust 0.35 l/(s m 2 ) Mechanical supply / exhaust 0.35 l/(s m 2 ) - - 0.041 2.13 0.97 0.68 Table 2 Description of building structure elements Building External structure area, m 2 A B elements C U-value, W/(m 2 K) External wall 609.0 1.4 0.22 0.22 Roof 398.4 0.9 0.11 0.11 Windows 249.4 27 1.7 1.4 Basement wall 146.0 1.4 0.31 0.31 Internal wall 402.1 4.4 4.4 4.2 Internal floor 4.2 4.2 3,0 Basement floor 3.0 3.0 4,4 The areas of the walls, floor and ceiling of apartment 9 and description what they are connected to are given in table 3. ISBN: 978-1-61804-132-6 53
Table 3 Structure elements surrounding apartment 9 Area Structure Connected to m 2 Floor Apartment 7 52,0 Ceiling Apartment 11 52,0 External wall Outdoor air 10,8 External wall Outdoor air 10,3 Internal wall Apartment 10 9,5 Internal wall Staircase 12,1 Internal wall Apartment 4 21,5 Window Outdoor air 9,8 The internal gains and their level of use were taken from Estonian efficiency regulation and they are given in table 4 [11]. It was presumed that the windows are always closed and also that there is uninsulated heating system piping in the apartments. Table 4 Description of internal gains Internal load People Lighting Equipment Max. heat gain, W/m 2 3 8 3 Average heat gain in relation to maximum, % 60 10 60 Energy simulation software IDA ICE 4.0 was used to calculate heat transmission between apartments and indoor air temperatures. This software allows the modelling of a multi-zone building, HVACsystems, internal and solar loads, outdoor climate, etc. and provides simultaneous dynamic simulation of heat transfer and air flows. It is a suitable tool for the simulation of thermal comfort, indoor air quality, and consumption in complex buildings. A modular simulation application, IDA ICE, has been developed by the Division of Building Services Engineering, KTH, and the Swedish Institute of Applied Mathematics, ITM [12]. IDA ICE has been tested against measurements [13] and several independent intermodel comparisons have been made [14]. In the comparisons, the performance of radiant heating and cooling systems using five simulation programs (CLIM2000, DOE, ESP-r, IDA-ICE and TRNSYS) were compared and IDA ICE showed a good agreement with the other programs. The tests and the comparisons showed a good justification for selecting IDA ICE as a reference tool in this study. Estonian test reference year was used in the simulations [15]. 3 Simulation results The consumption and indoor air temperatures of an apartment that was either heated or unheated were calculated. In addition, the consumptions of apartments surrounding the apartment and also the entire building were calculated. 3 different variants of an apartment building, which are described in the previous chapter, were analyzed. The heating specific consumptions before and after turning off the heating in apartment 9 are given in the form of tables, where the apartments (apt.) are situated similarly to the real situation. 3.1 A The heating specific consumptions of the surrounding apartments before the heating was turned off in apartment 9 are given in table 5 for variant A. The heat consumptions of apartments are compared to the weighted average of the building between 55% in the middle of the building and 126% on the top floor. A large part of the heat consumption is caused by heat loss through the building envelope and the areas of external structures of the apartments differ a lot, which thereby causes large variations in the specific heat consumptions of the apartments. Table 5 Specific heat consumptions of apartments when apartment 9 is heated (variant A) 6 259 11 258 12 258 126% 125% 126% 4 115 9 113 10 115 56% 55% 56% 2 138 7 135 8 138 67% 66% 67% The specific heat consumption of apartment 9 is 113 kwh/m2 a, which is 55% of the weighted average. The heat consumption of apartment 9 during the simulated period is 5858 kwh, which makes up 2.4% of the total consumption. The heat consumptions of the apartments adjacent to apartment 9, after the heating was turned off in it, are given in table 6. The rise in heat consumption is also given. It can be seen that the specific heat consumption of the adjacent apartments increased from 4.2% up to 19.0% ISBN: 978-1-61804-132-6 54
Table 6 Specific heat consumptions of apartments when apartment 9 is unheated (variant A) 6 261 (+2) 11 276 (+18) 12 260 (+2) +0.8% +7.0% +0.5% 4 126 (+11) 9 0 (-113) 10 120 (+5) +9.4% -100.0% +4.2% 2 140 (+2) 7 161 (+26) 8 139 (+1) +1.4% +19.0% +0.9% The heating consumption of apartment 9 5858 kwh - was reduced to zero, however the total heating consumption of the building was reduced only by 1596 kwh due to the increased heating of other apartments. The situation in apartment 9 is described in figure 2, where the cumulative indoor air temperatures of the heated and unheated apartment are given. It can be seen that the indoor temperature does t drop below +14.1 C throughout the year and it is below +18 C 3400 hours per year. Probably the inhabitants of the apartment would feel quite a lot of thermal discomfort and in real life heating would be occasionally turned on. The indoor temperatures of the adjacent apartments did t change significantly. Figure 2 Cumulative indoor temperatures in apartment 9 (variant A) 3.2 B The heating specific consumptions of the surrounding apartments before the heating was turned off in apartment 9 are given in table 7 for variant B. The heat consumptions of apartments are compared to the weighted average of the building between 88% in the middle of the building and 112% on the top floor. The fluctuations in the heating specific consumption of different apartments are smaller compared to variants A and C. The external envelope has been well insulated and a large part of the heat consumption is caused by unheated supply air. The air flow rate per m 2 of the apartments is constant and therefore specific heat losses are similar. Table 7 Specific heat consumptions of apartments when apartment 9 is heated (variant B) 6 64 11 63 12 64 112% 110% 111% 4 52 9 51 10 51 89% 88% 89% 2 58 7 57 8 58 The specific heat consumption of apartment 9 is 51 a, which is 88% of the weighted average. The heat consumption of apartment 9 during the simulated period is 2634 kwh, which makes up 3.9% of the total consumption. The heat consumption of the apartments adjacent to apartment 9, after the heating has been turned off in it, are given in table 8. The rise in heat consumption is also given. It can be seen that the specific heat consumption of the adjacent apartments has increased from 4.6% up to 23.5% Table 8 Specific heat consumptions of apartments when apartment 9 is unheated (variant B) 6 66 (+2) 11 75 (+12) 12 65 (+1) +1.8% +17.8% +1.1% 4 57 (+3) 9 0 (-51) 10 54 (+3) +9.7% -100.0% +4.6% 2 59 (+1) 7 70 (+13) 8 58 (+0) +1.8% +23.5% +1.2% The heating consumption of apartment 9 2634 kwh - was reduced to zero, however the total heating consumption of the building was reduced only by 375 kwh due to the increased heating of other apartments. The situation in apartment 9 is described in figure 3, where the cumulative indoor air temperatures of the heated and unheated apartment are given. It can be seen that the indoor temperature ISBN: 978-1-61804-132-6 55
does t drop below +17.2 C throughout the year and it is below +18 C only 200 hours per year. It means that thermal comfort of inhabitants can be achieved by adjusting clothing for most of the heating period. Figure 3 Cumulative indoor temperatures in apartment 9 (variant B) 3.2 C The heating specific consumptions of the surrounding apartments before the heating was turned off in apartment 9 are given in table 9 for variant C. The heat consumptions of apartments are compared to the weighted average of the building between 51% in the middle of the building and 107% on the top floor. It must be ted that corner apartments have the largest heating specific consumptions, but they are t shown in the tables. Besides the well insulated external envelope, the supply air is also preheated and that increases the fluctuations in specific heat consumptions compared to variant B as the weight of heat loss through the building envelope increases. Table 9 Specific heat consumptions of apartments when apartment 9 is heated (variant C) 6 30 11 29 12 29 107% 104% 107% 4 15 9 14 10 15 55% 51% 55% 2 20 7 19 8 20 73% 70% 74% The specific heat consumption of apartment 9 is 14 kwh/m2 a, which is 51% of the weighted average. The heat consumption of apartment 9 during the simulated period is 738 kwh, which makes up 2.2% of the total consumption. The heat consumption of the apartments adjacent to apartment 9, after the heating has been turned off in it, are given in table 10. The rise in the heat consumption is also given. It can be seen that the specific heat consumption of the adjacent apartments has increased from 3.9% up to 18.3% Table 10 Specific heat consumptions of apartments when apartment 9 is unheated (variant C) 6 30 11 32 12 30 +1.0% +10.8% +0.5% 4 16 9 0 10 16 +8.6% -100.0% +3.9% 2 21 7 23 8 20 +1.4% +18.3% +0.9% The heating consumption of apartment 9 738 kwh - was reduced to zero, however the total heating consumption of the building was reduced only by 149 kwh due to the increased heating of the other apartments. The situation in apartment 9 is described in figure 4, where the cumulative indoor air temperatures of the heated and unheated apartment are given. It can be seen that the indoor temperature does t drop below +19.0 C throughout the year. It means that thermal comfort of inhabitants can be achieved by adjusting clothing throughout the heating period. Figure 4 Cumulative indoor temperatures in apartment 9 (variant C) 4 Conclusion In conclusion it can be said that an unheated apartment has a significant effect on the heat consumption of the apartments surrounding it. In study 3 different variants of an apartment building ISBN: 978-1-61804-132-6 56
were analyzed 1 t refurbished one (A) and 2 refurbished buildings of which one did t have ventilation heat recovery (B) and the other did (C). The heat consumption of an apartment adjacent to an unheated one may increase up to 19.0%, 23.5% and 18.3% in the case of variants A, B and C respectively. The apartment, in which the heating was turned on, had annual heat consumptions of 5858 (A), 2634 (B) and 738 (C) kwh/year. However, the total heat consumption of the building was only reduced by1596, 375, and 149 kwh/year. This means that most of the heat needed to maintain indoor temperatures in an unheated apartment is obtained from the adjacent apartments. The situation in an unheated apartment is even better described by its indoor temperatures. The air temperatures during the heating period did t drop below +14.1 C, +17.2 C and +19.1 C in the case of variants A, B and C respectively. In the case of variants A and B the temperature was below +18 C for 3400 and 200 hours respectively. This means that only in case of a poorly insulated apartment building the heating system must be occasionally turned on so the inhabitants have the possibility of assuring thermal comfort by adjusting clothing. Only the heat consumption and indoor climate were studied in case the unheated apartment has only two external structure elements. Manipulating with the heating system of apartments with different locations is also needed. Besides it is common that there is uninsulated heating piping in the apartments, which can emit a significant amount of heat and by that increase indoor air temperatures. That is ather aspect that needs further study. References: [1] R. Kyrö, J. Heinen, A. Säynäjoki, S. Junnila, Occupants have little influence on overall consumption in district heated apartment buildings, Energy and Buildings, 43, 2011, pp. 3484-3490 [2] B.P. Xu, A. Huang, L. Fu, H.F. Di, Simulation and analysis on control effectiveness of TRVs in district heating systems, Energy and Buildings, 43, 2011, pp. 1169-1174 [3] B.P. Xu, A. Huang, L. Fu, H.F. Di, Dynamic simulation of space heating systems with radiators controlled by TRVs in buildings, Energy and Buildings, 40, 2008, pp. 1755-1764 [4] S. Karjalainen, Gender differences in thermal comfort and use of thermostats in everyday thermal environments, Building and Environment, 42, 2007, pp. 1594-1603 [5] S. Karjalainen, User problems with individual temperature control in offices, Building and Environment, 42, 2007, pp. 2880-2887 [6] J. Pakanen, S. Karjalainen, Estimating static heat flows in buildings for allocation systems, Energy and Buildings, 38, 2006, pp. 1044-1052 [7] C. Dimitroulopoulou, Ventilation in European dwelling: A review, Building and Environment, 47, 2012, pp. 109-125 [8] J.S. Park, H.J. Kim, A field study of occupant behavior and consumption in apartments with mechanical ventilation, Energy and Buildings, 50, 2012, pp. 19-25 [9] K.C. Parsons, The effects of gender, acclimation state, the opportunity to adjust clothing and physical disability on requirements for thermal comfort, Energy and Buildings, 34, 2002, pp. 593-599 [10] M. Frontczak, P. Wargocki, Literature survey on how different factors influence human comfort in indoor environments, Building and Environment, 46, 2011, pp. 922-937 [11] Estonian Government Regulation 258 Energiatõhususe miinimumnõuded, Riigiteataja, 2007 [12] P. Shalin, Modelling and simulation methods for modular continuous system in buildings, Doctoral Dissertation, KTH, Stockholm, 1996. [13] S. Moinard, G. Guyon (Eds.), Empirical Validation of EDF ETNA and GENEC Test- Cell Models, Subtask A.3, A Report of IEA Task 22, Building Energy Analysis Tools, 1999, 68 pp. [14] M. Achermann, G. Zweifel, RADTEST Radiant Heating and Cooling Test Cases, Subtask C, AReport of IEATask 22, Building Energy Analysis Tools, 2003, 83 pp. [15] T. Kalamees, J. Kurnitski, Estonian test reference year for calculations, In: Proceedings of the Estonian Academy of Sciences Engineering, 2006, 12, 1, pp. 40-58. ISBN: 978-1-61804-132-6 57