- Email: [email protected]
Conversion of electric heating in buildings
Energy and Buildings 40 (2008) 2188–2195
Contents lists available at ScienceDirect
Energy and Buildings journal homepage: www.elsevier.com/locate/enbuild
Conversion of electric heating in buildings An unconventional alternative Claes Blomqvist * University of Ga¨vle, Division of Indoor Environment, SE-801 76 Ga¨vle, Sweden
A R T I C L E I N F O
A B S T R A C T
Article history: Received 24 January 2008 Accepted 17 June 2008
To decrease the electric energy used for heating buildings it has become desirable to convert direct electrical heating to other heat sources. This paper reports on a study of the possibility of using an unconventional method for conversion to avoid installing an expensive hydronic system. The conversion method combines the ventilation and heating systems and uses air instead of water for distribution of heat within the building, taking advantage of thermal forces and the special properties of gravity currents. Full-scale tests have been carried out in a test apartment inside a laboratory hall where the conditions could be controlled. Temperatures and efficiency of ventilation have been measured to ensure that the demands with respect to thermal climate and air exchange were fulfilled. The results show that it is possible to use the method for heating and ventilation when converting the heating system, but further work has to be done to develop a detailed solution that works in practice. ß 2008 Elsevier B.V. All rights reserved.
Keywords: Thermal forces Large openings Gravity currents Electrical heating Conversion Heat transfer District heating
This work reports on an experimental study of the prerequisites of distributing heat using air as the heat carrier in order to avoid introducing an expensive hydronic heating system when converting direct electric heating to other heat sources.
the under-pressure caused by the ventilation system and entering the building through purpose-built openings usually located above or below the windows. During winter time, the temperature of the supply air often is very low; therefore, there are frequent complaints of cold draught in buildings equipped with stack or extract ventilation.
1.1. Background
1.2. Conventional conversion
Residential buildings in Sweden constructed during the 1960s and 1970s are often equipped with electric radiators for heating. The main reasons to choose electric heating were the low installation cost and the low price of electricity. Today the situation on the energy market is quite different and the price of electricity has increased significantly, an increase we can expect to continue in the future. Therefore it has become desirable to convert the heating system to reduce electricity usage. Examples of alternative energy sources are district heating or heat pumps. The most common ventilation principles in the actual buildings are stack ventilation or extract ventilation, where the air is transported from the building through openings in the kitchen and bathroom. Replacement air comes directly from outside, driven by
The conventional method to convert from electric to district heating is to install a hydronic system with a radiator below each window. This type of heating system is known to work satisfactorily. A disadvantage with a hydronic system is, however, the great installation cost.
1. Introduction
* Tel.: +46 26 648150. E-mail address: [email protected]. 0378-7788/$ – see front matter ß 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.enbuild.2008.06.012
1.3. Unconventional conversion of heating and ventilation An alternative solution is to use air as the heat carrier, thus combining the distribution of heat and ventilation air. An unconventional way to achieve this is to make use of so-called gravity currents for distribution of warm air for heating and ventilation. As the supply air will be heated before it enters the occupied zone it is likely that this solution also will reduce the risk of complaints from the occupants about cold draught from supply air inlets. To accomplish the unconventional conversion there are two phenomena of crucial importance:
C. Blomqvist / Energy and Buildings 40 (2008) 2188–2195
2189
Air movements caused by thermal forces The special properties of gravity currents with positive buoyancy (warm currents) There are also a number of questions that must be answered about the performance of the conversion method: 1. 2. 3. 4.
Are the thermal forces sufficient for the distribution of heat? Why should we use gravity currents? Can we meet the thermal comfort demands? Will the increased heat loss through the ceiling/roof be of importance? 5. Is air leakage from the building envelope of importance? 1.4. Air movements through large openings Thermal forces can be used to distribute heat and ventilation air within a building. Even very small temperature differences can cause large airflow rates through the doorway between adjacent rooms. If there are only thermal forces present the airflow pattern will consist of two flows of equal size and opposite direction (see Fig. 1). If the area of the opening is A ¼ h w, the flow rate in each direction can be written as: qe ¼ C A ðg 0 hÞ
1=2
(1)
where g0 is the reduced gravity (g0 = gDT/T) and C is a proportionality coefficient depending on several parameters which must be determined experimentally. Etheridge and Sandberg have summarised a number of studies carried out to determine C, both at full scale with air as the operating fluid and on a model scale using other operating fluids [1]. The value of the coefficient was found to vary between 0.12 and 0.20. Some of the studies reported that the coefficient showed a weak temperature dependence. Thermal-driven flows through horizontal openings can be described in a way similar to the flow through vertical openings [2,3]. However, the proportionality coefficient, C, is significantly less due to the more complex flow pattern (see Fig. 2). The conclusion is that the potential of distributing heat through horizontal openings using thermal forces is substantially less than for vertical openings and will therefore be excluded from this study. 1.5. Gravity currents A gravity current consists of a fluid propagating along a horizontal surface driven by its density difference to the ambient
Fig. 2. Airflow pattern in horizontal opening (T1 > T2).
fluid [4]. In ventilation applications the fluid is the ventilation air and the difference in density is caused by temperature difference. When cool air is introduced to a room at low momentum flux, a gravity current will develop along the floor and the distribution of air will be governed to a certain extent by the heat sources in the room. In a similar way a warm gravity current will spread along the ceiling and distribute heat to the parts of the room where the heat requirements are greatest. An interesting property of gravity currents is their ability to pass obstacles easily. Fig. 3 shows a series of pictures from a model test where a cool two-dimensional gravity current passes obstacles of various heights. For the visualisation the shadowgraph technique was used [5], with salt water as the supply fluid to achieve the density difference to the ambient. The inlet conditions correspond to a temperature difference of 3 8C on a full scale with air as the operating fluid. 1.5.1. Differences between cold and warm gravity currents 1.5.1.1. Cool gravity currents (negative buoyancy—ventilation). Cool gravity currents are used in connection with displacement ventilation. When cool air enters the room at low momentum, a gravity current will develop, distributing the supply air evenly over the floor while the heat sources in the room transport the air upwards into the occupied zone. Then the air is carried further to the extract devices, usually located close to the ceiling level. 1.5.1.2. Warm gravity currents (positive buoyancy—heating). In this study the building is to be heated using gravity currents propagating along the ceiling. Within the occupied zone there are no heat sinks corresponding to the heat sources as there are in the cooling case. The cold external wall and the window will, however, serve as heat sinks and contribute to the distribution of heat within the room. Whether this is enough to avoid an unacceptable thermal climate due to high temperature gradients or excessively low local temperatures has to be explored. 1.6. Unconventional heating and ventilation
Fig. 1. Airflow pattern in vertical opening (T1 > T2).
The aim of the present work is to explore the possibility of using thermal forces and gravity currents to distribute heat within a building. The warm air is supplied to the room at ceiling level by thermal forces from a neutral space, passes the occupied zone and leaves the room through ventilation openings at floor level (see Fig. 4). The airflow rate and consequently the amount of heat that is transferred are determined by the temperature difference between
2190
C. Blomqvist / Energy and Buildings 40 (2008) 2188–2195
Fig. 3. Visualisation of a cool gravity current passing an obstacle (water model).
Fig. 4. The principle of the unconventional distribution of heat from the hall (neutral space).
the supply and extract air and the effective area of the ventilation openings. This means that when the heating load is increased the temperature difference and the thermal forces will grow; the system will be self-controlling to a certain extent. A prerequisite to be able to carry out the unconventional conversion is that the building includes a room adjacent to all other rooms in the building. A building with those properties is said to have a neutral design and the central room is called a neutral space (see Fig. 4). An inventory of the Swedish building stock has shown that a large number of apartments have a hall serving as a neutral space that could be used for the distribution of warm air [6].
were carried out in a test apartment that was built inside the laboratory hall of our department. The apartment is a five-room building including kitchen, hall and bathroom (Fig. 5). In the test apartment the hall represents the neutral space mentioned above. As the intention was to carry out the measurements at steadystate conditions, we studied the bedroom and the hall, where full control of the thermal conditions could be obtained. In the living room the thermal climate was influenced by weather conditions and was therefore not possible to control completely. The external walls, floor and ceiling of the building are thermally insulated by fibreglass, while the internal walls are without thermal insulation. Table 1 shows the U-values of the test building.
2. Methods
2.2. Measurements of temperature profile in the bedroom
2.1. Test building
To study the temperature profile generated by warm air introduced to a room at ceiling level, an electric convector was placed on the rear wall of the bedroom (see Fig. 6). The power of the convector was varied between 150 W and 300 W, thus creating a gravity current moving along the ceiling towards the window. The space outside the bedroom window could be cooled down to 20 8C. To measure the temperature profile in the centre of the room, thermocouples were placed at 14 different levels (see Fig. 6). The surface temperatures of the floor and of the ceiling were also measured. Because of the low temperature outside the window we could expect a risk of thermal discomfort due to cold draught from the window. Therefore the flow pattern close to the window was studied using smoke visualisation and the air velocity was measured by means of a calibrated thermo-anemometer.
This is an experimental study of heat and air exchange between different parts of a multi-room building. Full scale measurements
Table 1 U-values of the test building (W/m2 K)
Fig. 5. Layout of the test apartment.
Ceiling External walls Floor Windows
0.19 0.35 0.20 2.0
C. Blomqvist / Energy and Buildings 40 (2008) 2188–2195
2191
Fig. 6. Test apartment. Bedroom with location of temperature measuring points. To the left is a cross-section and to the right is a plan view.
Since all tests were planned to be performed at steady-state conditions, the test building was allowed to stabilise for a period of 2–4 days. During this period of time the temperatures in the building were recorded every fifth minute. Through this procedure we could assure that steady-state conditions were obtained. Measurements were carried out for various heating power and ‘‘outdoor’’ temperatures. The test conditions for all test cases are listed in Table 2. 2.3. Heat transfer between hall and bedroom To study the heat distribution within the test apartment, electric convectors were placed on the upper part of the rear wall of the hall. The apartment was ventilated by an extract system, where the extract air was taken from kitchen and bathroom and air was supplied through an inlet device located in the rear part of the hall. Figs. 7 and 8 show the principle of heating and ventilation of the apartment during the tests. The exhaust airflow rates were constant, 10 l/s in the kitchen and 15 l/s in the bathroom, which means that the average nominal time constant of the test apartment was 2 h. Openings at ceiling and floor level in bedroom and living room provided ventilation air and heat to these rooms. The width of the openings was fixed at 0.55 m and the height was adjustable between 0.0 m and 0.25 m. The temperature profile was measured by thermocouples placed at different heights at the cross-marks in Fig. 7. To examine the influence of the size of the ventilation openings, measurements first were carried out when the heights of the openings were varied. During the main study the opening height was 0.1 m. For all tests during the main study the heating power
supplied was 500 W and the space outside the bedroom and kitchen windows (‘‘outdoors’’) was chilled to 18 8C. The air distribution, both in the hall and the bedroom of the test building, was also visualised and videotaped using smoke. The local mean age of air in a ventilated room is an indicator of the ventilation efficiency of the room. Therefore, the local mean age of air for some of the cases was measured using tracer gas at a constant flow rate, injected in the supply air (step up method) and measured in the centre of the five rooms. Measurements were recorded over a period of 12 h for each case (Fig. 8). In Table 3 the conditions during the tests carried out to examine the heat transfer from the hall to the bedroom are listed. Table 4 describes the test cases for the study of heat distribution within the test building. 3. Results 3.1. Measurements of temperature profile in the bedroom Examples of typical temperature profiles are presented in Fig. 9. The diagram shows that there is a layer of warm air of 10–15 cm thickness close to the ceiling. The vertical temperature gradient within the occupied zone, which is defined as 0.1–1.8 m above floor level, is approximately 1 8C. The air velocity of the cold draught from the window does not exceed 0.1 m/s.
Table 2 Conditions for the nine test cases Test_ID
Convector power (W)
Temperature outside window (8C)
1 2 3 4 5 6 7 8 9
250 200 200 250 300 300 250 200 150
15 5 15 15 15 20 5 +5 +5
Fig. 7. The air distribution within the test building during the tests.
2192
C. Blomqvist / Energy and Buildings 40 (2008) 2188–2195 Table 3 Conditions for the main tests of the heat distribution Heating power Temperature ‘‘outside’’ Ambient temperature Height of ventilation opening Extract air flow rate
500 W (heat source on rear wall of the hall) 18 8C (cooling chamber) +19 8C 0.10 m 25 l/s (0.5 ach/h)
Table 4 Description of test cases Case ID
Doors
Openings Upper
Lower
B D (reference case) E F G H
Closed Open Open Closed Closed Closed
Open Closed Open Open Open Open
Open Closed Closed Closed Closed Open
Fig. 8. Heating and ventilation of the bedroom.
In Fig. 10 the temperature profile for the nine cases (see Table 2) is plotted as the deviation from the average temperature in the occupied zone. We find that the shape of the profile is almost independent of the heating power and the outdoor temperature.
Fig. 9. Example of temperature profile in bedroom (no external ventilation).
Fig. 10. Deviation from average room temperature for the nine cases from Table 2.
C. Blomqvist / Energy and Buildings 40 (2008) 2188–2195
2193
The difference between the highest and lowest temperatures within the occupied zone is about 1 8C. 3.2. Heat transfer between hall and bedroom
Fig. 11. Temperature difference between rooms versus opening height (doors closed).
To evaluate the efficiency of the heat transfer from the hall to the bedroom, the average temperatures in the occupied zone between the rooms were compared. In order to determine a suitable height for the ventilation openings, tests were carried out for three different heights and closed doors. Fig. 11 shows the result of the measurements as the temperature differences between the two rooms versus the height of the openings. Since the effect of increasing the height to more than 0.10 m was small, the following tests were carried out for that height. Fig. 12 shows the temperature profiles in the hall and the bedroom for the case with open doors and closed ventilation slots. It is common in ordinary apartments that internal doors are left open. An open door provides maximum contact between the rooms, so this case has been used as the reference case.
Fig. 12. Measured temperature profiles in the hall and the bedroom.
Fig. 13. The result from the main study shown as the temperature difference between the hall and the bedroom.
2194
C. Blomqvist / Energy and Buildings 40 (2008) 2188–2195
Fig. 14. The warm air enters the bedroom and a gravity current is developed.
Fig. 13 shows the result of the measurements for all cases described in Table 4. The heat transfer resistance between the hall and the bedroom is expressed as the temperature difference between the two rooms. The temperatures are average temperatures in the occupied zone in the rooms. With closed doors the tests were repeated to get some insight into the accuracy. With the doors closed and only one opening open the heat transfer occurs mainly by conduction through the non-insulated wall between the rooms. Therefore the temperature difference with closed doors is about twice as the temperature difference when openings at both ceiling and floor levels exist. Fig. 14 shows a smoke visualisation of the air entering the bedroom of the test apartment and being distributed evenly along the ceiling as a 3D gravity current. During the visualisation the bedroom door was closed, but both upper and lower ventilation openings were open. 3.3. Tracer gas measurements The result of the constant flow tracer gas measurements of the local mean age of air is 2.5 h, to be compared with the nominal time constant of the apartment, which is 2.0 h. This means that we can consider the bedroom to be sufficiently well ventilated. 4. Discussion The introduction section listed a number of questions that I hoped this work would answer. Here each will be discussed, one by one. 1. Are the thermal forces sufficient for the distribution of heat? As a measure of the resistance between the source room (hall) and a neighbouring room, use the temperature difference between the hall and the rooms. The difference amounted to about 2 8C if we use the average temperature in the occupied zone for the calculation. The main part of the heat power is transferred by the warm air and the remaining part through conduction through the non-insulated internal wall. The heating power supplied by the warm air can be increased by making the ventilation openings wider. The conclusion is that the measured temperature differences would be acceptable by the occupants. 2. Why should we use gravity currents? The gravity current will distribute the air and subsequently the heat by its temperature difference to the parts of the room where the demand is the greatest. The supply of heat is therefore self-regulating. The
obstacles in the way of the current will easily be passed and the air will be evenly distributed in the room. 3. Can we meet the thermal comfort demands? The most common reasons for thermal discomfort are local cooling of the body caused by high air speed, low temperatures and large vertical temperature gradients. The experiments have shown that the air velocities in the occupied zone do not exceed 0.10 m/s, which meets the comfort criteria. The temperature gradient within the occupied zone can also be considered as acceptable. 4. Will the increased heat loss through the ceiling/roof be of importance? There will be increased heat loss through the ceiling due to the high temperature of the air close to the ceiling, but the aim of the unconventional conversion is to lower the need for high-quality electricity and replace it with low-temperature heat sources such as district heating. As an example, district heating can use renewable material as fuel and therefore it can be justifiable to accept a small increase in heat loss through the ceiling. One can also notice that many of the houses built during the 1960s and 1970s are in need of refurbishing; this conversion would provide a suitable opportunity to improve the thermal insulation of the ceiling, as well. 5. Is air leakage from the building envelope of importance? It is likely that the actual houses are rather leaky which may cause complaints from the occupants of cold draught leaking in directly from outside through cracks in the external walls. As mentioned above, the actual buildings are in need of remodeling, and during which the airtightness of the building could also be improved. The conversion method has a potential to work from a thermal point of view, but there are a number of details that must be settled before it could be used in practice. To study the performance of the conversion method it would be preferable to carry out tests on a real building under realistic conditions during a winter season. Sound transmission through the relatively large ventilation openings between the rooms may cause inconvenience; measures must probably be taken to reduce such problems without increasing the flow resistance in the ventilation openings. This work was focused on the distribution of heat and air in the building, so water heating was not taken into consideration. Equipment for this that uses heat for example from a district heating network also has to be developed before the conversion method can be used in practice. An additional problem to solve is heating rooms such as the bathroom and kitchen, from which we do not want to receive
C. Blomqvist / Energy and Buildings 40 (2008) 2188–2195
recirculated air. Future studies could incorporate numeric simulations and computational fluid dynamics. Acknowledgment The author would like to thank Professor Mats Sandberg and Professor Tor-Go¨ran Malmstro¨m for valuable comments and fruitful discussions. Financial support from Formas (BIC 2), Ga¨vle Energi AB, Falu Energi och Vatten AB, Borla¨nge Energi AB and Sandviken Energi AB is gratefully acknowledged.
2195
References [1] D. Etheridge, M. Sandberg, Building Ventilation, Theory and Measurement, John Wiley & Sons, Chichester, UK, 1996, pp. 508–509. [2] C. Blomqvist, M. Sandberg, Air movements through horizontal openings in buildings, International Journal of Ventilation 3 (1) (2004) 1–9. [3] P. Heiselberg, Z. Li, Experimental study of buoyancy driven natural ventilation through horizontal openings, in: Proceedings of ROOMVENT, vol. 2. Helsinki, Finland, 2007, pp. 141–150. [4] J.E. Simpson, Gravity Currents, In the Environment and the Laboratory, Ellis Horwood Ltd., 1987. [5] G.S. Settles, Schlieren and Shadowgraph Techniques, Springer Verlag, 2001. [6] C. Blomqvist, M. Mattsson, et al., Grundla¨ggande fo¨rutsa¨ttning fo¨r konvertering av direkteluppva¨rmda hus till fja¨rrva¨rme genom okonventionell placering av va¨rmeka¨llor (Report in Swedish), 2000.
Contents lists available at ScienceDirect
Energy and Buildings journal homepage: www.elsevier.com/locate/enbuild
Conversion of electric heating in buildings An unconventional alternative Claes Blomqvist * University of Ga¨vle, Division of Indoor Environment, SE-801 76 Ga¨vle, Sweden
A R T I C L E I N F O
A B S T R A C T
Article history: Received 24 January 2008 Accepted 17 June 2008
To decrease the electric energy used for heating buildings it has become desirable to convert direct electrical heating to other heat sources. This paper reports on a study of the possibility of using an unconventional method for conversion to avoid installing an expensive hydronic system. The conversion method combines the ventilation and heating systems and uses air instead of water for distribution of heat within the building, taking advantage of thermal forces and the special properties of gravity currents. Full-scale tests have been carried out in a test apartment inside a laboratory hall where the conditions could be controlled. Temperatures and efficiency of ventilation have been measured to ensure that the demands with respect to thermal climate and air exchange were fulfilled. The results show that it is possible to use the method for heating and ventilation when converting the heating system, but further work has to be done to develop a detailed solution that works in practice. ß 2008 Elsevier B.V. All rights reserved.
Keywords: Thermal forces Large openings Gravity currents Electrical heating Conversion Heat transfer District heating
This work reports on an experimental study of the prerequisites of distributing heat using air as the heat carrier in order to avoid introducing an expensive hydronic heating system when converting direct electric heating to other heat sources.
the under-pressure caused by the ventilation system and entering the building through purpose-built openings usually located above or below the windows. During winter time, the temperature of the supply air often is very low; therefore, there are frequent complaints of cold draught in buildings equipped with stack or extract ventilation.
1.1. Background
1.2. Conventional conversion
Residential buildings in Sweden constructed during the 1960s and 1970s are often equipped with electric radiators for heating. The main reasons to choose electric heating were the low installation cost and the low price of electricity. Today the situation on the energy market is quite different and the price of electricity has increased significantly, an increase we can expect to continue in the future. Therefore it has become desirable to convert the heating system to reduce electricity usage. Examples of alternative energy sources are district heating or heat pumps. The most common ventilation principles in the actual buildings are stack ventilation or extract ventilation, where the air is transported from the building through openings in the kitchen and bathroom. Replacement air comes directly from outside, driven by
The conventional method to convert from electric to district heating is to install a hydronic system with a radiator below each window. This type of heating system is known to work satisfactorily. A disadvantage with a hydronic system is, however, the great installation cost.
1. Introduction
* Tel.: +46 26 648150. E-mail address: [email protected]. 0378-7788/$ – see front matter ß 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.enbuild.2008.06.012
1.3. Unconventional conversion of heating and ventilation An alternative solution is to use air as the heat carrier, thus combining the distribution of heat and ventilation air. An unconventional way to achieve this is to make use of so-called gravity currents for distribution of warm air for heating and ventilation. As the supply air will be heated before it enters the occupied zone it is likely that this solution also will reduce the risk of complaints from the occupants about cold draught from supply air inlets. To accomplish the unconventional conversion there are two phenomena of crucial importance:
C. Blomqvist / Energy and Buildings 40 (2008) 2188–2195
2189
Air movements caused by thermal forces The special properties of gravity currents with positive buoyancy (warm currents) There are also a number of questions that must be answered about the performance of the conversion method: 1. 2. 3. 4.
Are the thermal forces sufficient for the distribution of heat? Why should we use gravity currents? Can we meet the thermal comfort demands? Will the increased heat loss through the ceiling/roof be of importance? 5. Is air leakage from the building envelope of importance? 1.4. Air movements through large openings Thermal forces can be used to distribute heat and ventilation air within a building. Even very small temperature differences can cause large airflow rates through the doorway between adjacent rooms. If there are only thermal forces present the airflow pattern will consist of two flows of equal size and opposite direction (see Fig. 1). If the area of the opening is A ¼ h w, the flow rate in each direction can be written as: qe ¼ C A ðg 0 hÞ
1=2
(1)
where g0 is the reduced gravity (g0 = gDT/T) and C is a proportionality coefficient depending on several parameters which must be determined experimentally. Etheridge and Sandberg have summarised a number of studies carried out to determine C, both at full scale with air as the operating fluid and on a model scale using other operating fluids [1]. The value of the coefficient was found to vary between 0.12 and 0.20. Some of the studies reported that the coefficient showed a weak temperature dependence. Thermal-driven flows through horizontal openings can be described in a way similar to the flow through vertical openings [2,3]. However, the proportionality coefficient, C, is significantly less due to the more complex flow pattern (see Fig. 2). The conclusion is that the potential of distributing heat through horizontal openings using thermal forces is substantially less than for vertical openings and will therefore be excluded from this study. 1.5. Gravity currents A gravity current consists of a fluid propagating along a horizontal surface driven by its density difference to the ambient
Fig. 2. Airflow pattern in horizontal opening (T1 > T2).
fluid [4]. In ventilation applications the fluid is the ventilation air and the difference in density is caused by temperature difference. When cool air is introduced to a room at low momentum flux, a gravity current will develop along the floor and the distribution of air will be governed to a certain extent by the heat sources in the room. In a similar way a warm gravity current will spread along the ceiling and distribute heat to the parts of the room where the heat requirements are greatest. An interesting property of gravity currents is their ability to pass obstacles easily. Fig. 3 shows a series of pictures from a model test where a cool two-dimensional gravity current passes obstacles of various heights. For the visualisation the shadowgraph technique was used [5], with salt water as the supply fluid to achieve the density difference to the ambient. The inlet conditions correspond to a temperature difference of 3 8C on a full scale with air as the operating fluid. 1.5.1. Differences between cold and warm gravity currents 1.5.1.1. Cool gravity currents (negative buoyancy—ventilation). Cool gravity currents are used in connection with displacement ventilation. When cool air enters the room at low momentum, a gravity current will develop, distributing the supply air evenly over the floor while the heat sources in the room transport the air upwards into the occupied zone. Then the air is carried further to the extract devices, usually located close to the ceiling level. 1.5.1.2. Warm gravity currents (positive buoyancy—heating). In this study the building is to be heated using gravity currents propagating along the ceiling. Within the occupied zone there are no heat sinks corresponding to the heat sources as there are in the cooling case. The cold external wall and the window will, however, serve as heat sinks and contribute to the distribution of heat within the room. Whether this is enough to avoid an unacceptable thermal climate due to high temperature gradients or excessively low local temperatures has to be explored. 1.6. Unconventional heating and ventilation
Fig. 1. Airflow pattern in vertical opening (T1 > T2).
The aim of the present work is to explore the possibility of using thermal forces and gravity currents to distribute heat within a building. The warm air is supplied to the room at ceiling level by thermal forces from a neutral space, passes the occupied zone and leaves the room through ventilation openings at floor level (see Fig. 4). The airflow rate and consequently the amount of heat that is transferred are determined by the temperature difference between
2190
C. Blomqvist / Energy and Buildings 40 (2008) 2188–2195
Fig. 3. Visualisation of a cool gravity current passing an obstacle (water model).
Fig. 4. The principle of the unconventional distribution of heat from the hall (neutral space).
the supply and extract air and the effective area of the ventilation openings. This means that when the heating load is increased the temperature difference and the thermal forces will grow; the system will be self-controlling to a certain extent. A prerequisite to be able to carry out the unconventional conversion is that the building includes a room adjacent to all other rooms in the building. A building with those properties is said to have a neutral design and the central room is called a neutral space (see Fig. 4). An inventory of the Swedish building stock has shown that a large number of apartments have a hall serving as a neutral space that could be used for the distribution of warm air [6].
were carried out in a test apartment that was built inside the laboratory hall of our department. The apartment is a five-room building including kitchen, hall and bathroom (Fig. 5). In the test apartment the hall represents the neutral space mentioned above. As the intention was to carry out the measurements at steadystate conditions, we studied the bedroom and the hall, where full control of the thermal conditions could be obtained. In the living room the thermal climate was influenced by weather conditions and was therefore not possible to control completely. The external walls, floor and ceiling of the building are thermally insulated by fibreglass, while the internal walls are without thermal insulation. Table 1 shows the U-values of the test building.
2. Methods
2.2. Measurements of temperature profile in the bedroom
2.1. Test building
To study the temperature profile generated by warm air introduced to a room at ceiling level, an electric convector was placed on the rear wall of the bedroom (see Fig. 6). The power of the convector was varied between 150 W and 300 W, thus creating a gravity current moving along the ceiling towards the window. The space outside the bedroom window could be cooled down to 20 8C. To measure the temperature profile in the centre of the room, thermocouples were placed at 14 different levels (see Fig. 6). The surface temperatures of the floor and of the ceiling were also measured. Because of the low temperature outside the window we could expect a risk of thermal discomfort due to cold draught from the window. Therefore the flow pattern close to the window was studied using smoke visualisation and the air velocity was measured by means of a calibrated thermo-anemometer.
This is an experimental study of heat and air exchange between different parts of a multi-room building. Full scale measurements
Table 1 U-values of the test building (W/m2 K)
Fig. 5. Layout of the test apartment.
Ceiling External walls Floor Windows
0.19 0.35 0.20 2.0
C. Blomqvist / Energy and Buildings 40 (2008) 2188–2195
2191
Fig. 6. Test apartment. Bedroom with location of temperature measuring points. To the left is a cross-section and to the right is a plan view.
Since all tests were planned to be performed at steady-state conditions, the test building was allowed to stabilise for a period of 2–4 days. During this period of time the temperatures in the building were recorded every fifth minute. Through this procedure we could assure that steady-state conditions were obtained. Measurements were carried out for various heating power and ‘‘outdoor’’ temperatures. The test conditions for all test cases are listed in Table 2. 2.3. Heat transfer between hall and bedroom To study the heat distribution within the test apartment, electric convectors were placed on the upper part of the rear wall of the hall. The apartment was ventilated by an extract system, where the extract air was taken from kitchen and bathroom and air was supplied through an inlet device located in the rear part of the hall. Figs. 7 and 8 show the principle of heating and ventilation of the apartment during the tests. The exhaust airflow rates were constant, 10 l/s in the kitchen and 15 l/s in the bathroom, which means that the average nominal time constant of the test apartment was 2 h. Openings at ceiling and floor level in bedroom and living room provided ventilation air and heat to these rooms. The width of the openings was fixed at 0.55 m and the height was adjustable between 0.0 m and 0.25 m. The temperature profile was measured by thermocouples placed at different heights at the cross-marks in Fig. 7. To examine the influence of the size of the ventilation openings, measurements first were carried out when the heights of the openings were varied. During the main study the opening height was 0.1 m. For all tests during the main study the heating power
supplied was 500 W and the space outside the bedroom and kitchen windows (‘‘outdoors’’) was chilled to 18 8C. The air distribution, both in the hall and the bedroom of the test building, was also visualised and videotaped using smoke. The local mean age of air in a ventilated room is an indicator of the ventilation efficiency of the room. Therefore, the local mean age of air for some of the cases was measured using tracer gas at a constant flow rate, injected in the supply air (step up method) and measured in the centre of the five rooms. Measurements were recorded over a period of 12 h for each case (Fig. 8). In Table 3 the conditions during the tests carried out to examine the heat transfer from the hall to the bedroom are listed. Table 4 describes the test cases for the study of heat distribution within the test building. 3. Results 3.1. Measurements of temperature profile in the bedroom Examples of typical temperature profiles are presented in Fig. 9. The diagram shows that there is a layer of warm air of 10–15 cm thickness close to the ceiling. The vertical temperature gradient within the occupied zone, which is defined as 0.1–1.8 m above floor level, is approximately 1 8C. The air velocity of the cold draught from the window does not exceed 0.1 m/s.
Table 2 Conditions for the nine test cases Test_ID
Convector power (W)
Temperature outside window (8C)
1 2 3 4 5 6 7 8 9
250 200 200 250 300 300 250 200 150
15 5 15 15 15 20 5 +5 +5
Fig. 7. The air distribution within the test building during the tests.
2192
C. Blomqvist / Energy and Buildings 40 (2008) 2188–2195 Table 3 Conditions for the main tests of the heat distribution Heating power Temperature ‘‘outside’’ Ambient temperature Height of ventilation opening Extract air flow rate
500 W (heat source on rear wall of the hall) 18 8C (cooling chamber) +19 8C 0.10 m 25 l/s (0.5 ach/h)
Table 4 Description of test cases Case ID
Doors
Openings Upper
Lower
B D (reference case) E F G H
Closed Open Open Closed Closed Closed
Open Closed Open Open Open Open
Open Closed Closed Closed Closed Open
Fig. 8. Heating and ventilation of the bedroom.
In Fig. 10 the temperature profile for the nine cases (see Table 2) is plotted as the deviation from the average temperature in the occupied zone. We find that the shape of the profile is almost independent of the heating power and the outdoor temperature.
Fig. 9. Example of temperature profile in bedroom (no external ventilation).
Fig. 10. Deviation from average room temperature for the nine cases from Table 2.
C. Blomqvist / Energy and Buildings 40 (2008) 2188–2195
2193
The difference between the highest and lowest temperatures within the occupied zone is about 1 8C. 3.2. Heat transfer between hall and bedroom
Fig. 11. Temperature difference between rooms versus opening height (doors closed).
To evaluate the efficiency of the heat transfer from the hall to the bedroom, the average temperatures in the occupied zone between the rooms were compared. In order to determine a suitable height for the ventilation openings, tests were carried out for three different heights and closed doors. Fig. 11 shows the result of the measurements as the temperature differences between the two rooms versus the height of the openings. Since the effect of increasing the height to more than 0.10 m was small, the following tests were carried out for that height. Fig. 12 shows the temperature profiles in the hall and the bedroom for the case with open doors and closed ventilation slots. It is common in ordinary apartments that internal doors are left open. An open door provides maximum contact between the rooms, so this case has been used as the reference case.
Fig. 12. Measured temperature profiles in the hall and the bedroom.
Fig. 13. The result from the main study shown as the temperature difference between the hall and the bedroom.
2194
C. Blomqvist / Energy and Buildings 40 (2008) 2188–2195
Fig. 14. The warm air enters the bedroom and a gravity current is developed.
Fig. 13 shows the result of the measurements for all cases described in Table 4. The heat transfer resistance between the hall and the bedroom is expressed as the temperature difference between the two rooms. The temperatures are average temperatures in the occupied zone in the rooms. With closed doors the tests were repeated to get some insight into the accuracy. With the doors closed and only one opening open the heat transfer occurs mainly by conduction through the non-insulated wall between the rooms. Therefore the temperature difference with closed doors is about twice as the temperature difference when openings at both ceiling and floor levels exist. Fig. 14 shows a smoke visualisation of the air entering the bedroom of the test apartment and being distributed evenly along the ceiling as a 3D gravity current. During the visualisation the bedroom door was closed, but both upper and lower ventilation openings were open. 3.3. Tracer gas measurements The result of the constant flow tracer gas measurements of the local mean age of air is 2.5 h, to be compared with the nominal time constant of the apartment, which is 2.0 h. This means that we can consider the bedroom to be sufficiently well ventilated. 4. Discussion The introduction section listed a number of questions that I hoped this work would answer. Here each will be discussed, one by one. 1. Are the thermal forces sufficient for the distribution of heat? As a measure of the resistance between the source room (hall) and a neighbouring room, use the temperature difference between the hall and the rooms. The difference amounted to about 2 8C if we use the average temperature in the occupied zone for the calculation. The main part of the heat power is transferred by the warm air and the remaining part through conduction through the non-insulated internal wall. The heating power supplied by the warm air can be increased by making the ventilation openings wider. The conclusion is that the measured temperature differences would be acceptable by the occupants. 2. Why should we use gravity currents? The gravity current will distribute the air and subsequently the heat by its temperature difference to the parts of the room where the demand is the greatest. The supply of heat is therefore self-regulating. The
obstacles in the way of the current will easily be passed and the air will be evenly distributed in the room. 3. Can we meet the thermal comfort demands? The most common reasons for thermal discomfort are local cooling of the body caused by high air speed, low temperatures and large vertical temperature gradients. The experiments have shown that the air velocities in the occupied zone do not exceed 0.10 m/s, which meets the comfort criteria. The temperature gradient within the occupied zone can also be considered as acceptable. 4. Will the increased heat loss through the ceiling/roof be of importance? There will be increased heat loss through the ceiling due to the high temperature of the air close to the ceiling, but the aim of the unconventional conversion is to lower the need for high-quality electricity and replace it with low-temperature heat sources such as district heating. As an example, district heating can use renewable material as fuel and therefore it can be justifiable to accept a small increase in heat loss through the ceiling. One can also notice that many of the houses built during the 1960s and 1970s are in need of refurbishing; this conversion would provide a suitable opportunity to improve the thermal insulation of the ceiling, as well. 5. Is air leakage from the building envelope of importance? It is likely that the actual houses are rather leaky which may cause complaints from the occupants of cold draught leaking in directly from outside through cracks in the external walls. As mentioned above, the actual buildings are in need of remodeling, and during which the airtightness of the building could also be improved. The conversion method has a potential to work from a thermal point of view, but there are a number of details that must be settled before it could be used in practice. To study the performance of the conversion method it would be preferable to carry out tests on a real building under realistic conditions during a winter season. Sound transmission through the relatively large ventilation openings between the rooms may cause inconvenience; measures must probably be taken to reduce such problems without increasing the flow resistance in the ventilation openings. This work was focused on the distribution of heat and air in the building, so water heating was not taken into consideration. Equipment for this that uses heat for example from a district heating network also has to be developed before the conversion method can be used in practice. An additional problem to solve is heating rooms such as the bathroom and kitchen, from which we do not want to receive
C. Blomqvist / Energy and Buildings 40 (2008) 2188–2195
recirculated air. Future studies could incorporate numeric simulations and computational fluid dynamics. Acknowledgment The author would like to thank Professor Mats Sandberg and Professor Tor-Go¨ran Malmstro¨m for valuable comments and fruitful discussions. Financial support from Formas (BIC 2), Ga¨vle Energi AB, Falu Energi och Vatten AB, Borla¨nge Energi AB and Sandviken Energi AB is gratefully acknowledged.
2195
References [1] D. Etheridge, M. Sandberg, Building Ventilation, Theory and Measurement, John Wiley & Sons, Chichester, UK, 1996, pp. 508–509. [2] C. Blomqvist, M. Sandberg, Air movements through horizontal openings in buildings, International Journal of Ventilation 3 (1) (2004) 1–9. [3] P. Heiselberg, Z. Li, Experimental study of buoyancy driven natural ventilation through horizontal openings, in: Proceedings of ROOMVENT, vol. 2. Helsinki, Finland, 2007, pp. 141–150. [4] J.E. Simpson, Gravity Currents, In the Environment and the Laboratory, Ellis Horwood Ltd., 1987. [5] G.S. Settles, Schlieren and Shadowgraph Techniques, Springer Verlag, 2001. [6] C. Blomqvist, M. Mattsson, et al., Grundla¨ggande fo¨rutsa¨ttning fo¨r konvertering av direkteluppva¨rmda hus till fja¨rrva¨rme genom okonventionell placering av va¨rmeka¨llor (Report in Swedish), 2000.