Comparison between a radiant floor and two radiant walls on heating and cooling energy demand.
Radiant systems became very popular in the last decades. They are commonly used as radiant floor systems, although from the 1990s radiant ceilings have been increasingly installed. In the last years several producers propose radiant wall systems: these systems have some inconvenient when hanging pictures, as well as problems with furniture. Wall systems have an overall heat exchange coefficient equal to 8 W/([m.sup.2]*K) (1.41 Btu/[h*[ft.sup.2]*[degrees]F]) both in heating and cooling conditions (CEN 2008). Radiant floor systems have an overall heat exchange coefficient equal to 11 W/([m.sup.2]*K) (1.94 Btu/(h*[ft.sup.2]*[degrees]F)) in heating and equal to 7 W/(m*[.sup.2]K) (1.23 Btu/(h*[ft.sup.2]*[degrees]F)) in cooling conditions.
Many times manufacturers of radiant wall systems propose their application directly on external walls. In this case insulation behind pipes has to be applied, since, otherwise, losses become very high.
Since it does not seem clear the entity of the losses when applying a radiant floor or a radiant wall heating, in this work a comparison between those systems has been carried out; moreover the influence of position of insulation layer in radiant wall systems has been analyzed.
A comparison between different systems is almost impossible to carry out by means of measurements, since many uncertainties appear when measuring energy performance in buildings. Therefore the recourse to a dynamic simulation with detailed models seems inevitable when comparing different systems under the same conditions.
A suitable model should take into account the internal loads, solar radiation, as well as thermal conduction in transient conditions inside structures where pipes are embedded. Many models have been proposed in the last years. In this work the model DigiThon (Brunello et al. 2001) has been used. Such a model is based on transfer function method (Kusuda 1969, De Carli 2002), and it allows simulating whichever radiant system, since transfer functions can be calculated via two-dimension detailed model (Blomberg 1999).
In this way it is possible to evaluate hour by hour the energy demand of the radiant wall, both as useful thermal energy for the house and losses behind pipes.
In Figure 1 the plan of a floor of the case study is shown. It is a residential house, with two levels and the same room distribution. On the lower side, the ground floor is adjacent to no heated rooms where 10[degrees]C (50[degrees]F) air temperature has been considered for each considered climate.
[FIGURE 1 OMITTED]
All rooms have a radiant system (floor or wall depending on the considered case) except bathrooms and corridors with convective systems. In Figure 2 the thermal properties of the structures of the building are reported. The solar factor (ratio of the total solar energy flux entering through the glass to the incident solar energy flux) of the glazed surfaces is 0.35.
[FIGURE 2 OMITTED]
A mechanical ventilation system (0.5 Vol/h of air change per hour (ach)) with 50% sensible heat recovery has been considered: it runs always in no stop mode.
The considered radiant systems are shown in Figure 3. The radiant floor shown in Figure 3.I is the one used when the lower adjacent room is not heated, i.e. with additional insulation, as required by the Standard 1264 part 4 (CEN 2001), in order to reduce losses. In the slab between the first and second level, the insulation layer thickness becomes 3 cm (0.0984 ft).
[FIGURE 3 OMITTED]
Two radiant walls have been considered (Figure 3.II):
* case a: insulation layer is put on the inside and also on the outside part of the wall;
* case b: whole insulation layer is put on the outside part of the wall.
Three cities have been taken into account: Wuerzburg, Venice and Naples. In Figure 4 mean monthly days outdoor temperatures are shown and in Figure 5 beam and diffuse solar radiation are shown.
[FIGURE 4 OMITTED]
[FIGURE 5 OMITTED]
The radiant system is installed on the whole surface of floor in the case of radiant floor but when it is installed on the external walls the radiant surface is about 40% of floor surface: it depends on the zone load and difference between supply water temperature and room air. Only in Wuerzburg radiant walls have a greater surface, about 50% of the floor surface, since the outdoor air temperature is lower than other cities.
WINTER DESIGN DAY
In this study two different types of simulations have been carried out: winter design day simulations using data in Table 1 and simulations on the long period based on the test reference year (TRY). In winter design day simulations, solar radiation and internal loads have not been considered but in the long period analysis they have been taken into account.
Table 1. Winter Design Temperatures T Latitude Altitude Location [[degrees]C] ([degrees]F) [[degrees]] [m] (ft) Wuerzburg -15 (5) 49.0 182 (597) Venice -5 (23) 45.0 10 (33) Naples 2 (35.6) 40.0 17 (56)
Winter design day simulations allow to evaluate the required power and also supply water temperature to radiant systems compatible with comfort conditions. The power demand on the building side does not take into account convective systems of bathrooms and corridors: their power is almost equal for each case. Both with radiant floor and with radiant wall, operative temperature in the rooms is about 20[degrees]C (68[degrees]F).
The supply water temperature in the radiant system is shown in Table 2 for the three different climates. In winter design day, radiant systems run without control for the whole day (continuous running): these simulations are in steady state conditions. As it can be seen, design water supply temperature is rather high (10[degrees]C to almost 20[degrees]C [50[degrees]F to almost 68[degrees]F]) for radiant wall in comparison with radiant floor system. In Table 3 the total power and losses (ratio of radiant useful energy on the inside surface and overall water energy in the pipes) are presented for the three cities: the volume of insulation material is the same for radiant floor and radiant wall, therefore the power demand is almost the same. In Wuerzburg with the same water energy, the radiant floor energy is higher than radiant wall; in the other climates the ' systems are nearly the same. In steady state conditions, case a and case b radiant walls present no differences.
Table 2. Supply Water to Radiant Systems in Winter Design Day Radiant Floor Radiant Wall Location Supply T [DELTA] T Supply T [DELTA] T water water water water [[degrees]C] [[degrees]C] [[degrees]C] [[degrees]C] ([degrees]F) ([degrees]F) ([degrees]F) ([degrees]F) Wuerzburg 27 (80.6) 4 (7.2) 45 (113) 4 (7.2) Venice 25 (77.0) 4 (7.2) 45 (113) 4 (7.2) Naples 24 (75.2) 4 (7.2) 35 (95) 4 (7.2) Table 3. Power and Losses For Radiant Systems in Winter Design Day Radiant Floor Radiant Wall Radiant Floor Radiant Wall Power (1) Power (1) Losses (2) Losses (2) case a case b Location [kW] (Btu/h) [kW] (Btu/h) [-] [-] [-] Wuerzburg 9.1 (31047) 9.1 (31047) 0.87 0.76 0.76 Venice 6.3 (21494) 6.6 (22518) 0.84 0.85 0.85 Naples 5.0 (17059) 5.5 (18765) 0.81 0.85 0.85 (1) without taking into account convective power of bathrooms and corridors. (2) ratio between radiant useful energy on the inside surface and water energy at pipes level.
As for the power installed, Wuerzburg has the same values for both two systems, while in warmer climates the heating capacity of radiant floor is slightly lower than radiant wall.
ANNUAL HEATING ENERGY DEMAND
In TRY simulations an on/off control on the indoor air temperature with set-point of 20[degrees]C (68[degrees]F) has been considered.
In Figure 6 the specific yearly need of heating energy of the considered building is reported. In this case air to water heat pump has been taken into account, by means of an appropriate simulation tool (Scarpa 2007). In Table 4 mean COP and the amount of working hours of air to water heat pump in heating season are shown.
Table 4. Mean COP and Working Hours of Air to Water Heat Pump in Heating. Radiant Floor Radiant Wall case a case b Location COP Working hours COP Working hours COP Working hours [-] [-] [-] Wuerzburg 3.12 2797 2.84 4056 2.87 3998 Venice 3.23 2451 2.88 3427 2.98 3281 Naples 3.39 1545 2.89 2112 2.95 1977
[FIGURE 6 OMITTED]
The efficiency of the electricity production mix, to convert electricity into primary energy consumption, 37% has been considered. In Figure 7 the annual primary energy demand per net surface (259 [m.sup.2] [2788 [ft.sup.2]]) is shown (the area of bathrooms and corridors has not been considered): it takes into account also pumping energy, on the building side, with pump efficiency of 0.4. In Table 5 the seasonal mean supply and return water temperatures are shown. Radiant wall supply temperature is always higher than radiant floor: this is due to less surface and heat exchange efficiency. It is interesting to see that, in case b, the water supply temperature is slightly lower than case a.
Table 5. Seasonal Mean Supply and Return Water Temperatures for Radiant Systems in Heating Radiant Floor Location T input T output [[degrees]C] [[degrees]C] ([degrees]F) ([degrees]F) Wuerzburg 25.9 (78.6) 23.2 (73.8) Venice 24.7 (76.5) 22.5 (72.5) Naples 24.5 (76.1) 22.3 (72.1) Radiant Wall case a case b Location T input T output T input T output [[degrees]C] [[degrees]C] [[degrees]C] [[degrees]C] ([degrees]F) ([degrees]F) ([degrees]F) ([degrees]F) Wuerzburg 28.8 (83.8) 26.5 (79.7) 28.3 (82.9) 25.9 (78.6) Venice 29.8 (85.6) 27.3 (81.1) 29.4 (84.9) 26.8 (80.2) Naples 27.8 (82.0) 25.7 (78.3) 27.5 (81.5) 25.2 (77.4)
[FIGURE 7 OMITTED]
In Table 6 the ratio of radiant useful energy on the inside surface and overall water energy is reported. In Naples the radiant wall behaves better than radiant floor: this is probably due to solar radiation (for this city is higher than the other locations), which increases the temperature on external surfaces, thus reducing losses.
Table 6. Losses (1) of Radiant Systems in Heating Mode Radiant Floor Radiant Wall Location case a case b [-] [-] [-] Wuerzburg 0.75 0.71 0.67 Venice 0.71 0.79 0.76 Naples 0.65 0.78 0.72 (1) ratio between radiant useful energy on the inside surface and water energy at pipes level.
In order to obtain in case b the same indoor conditions of case a (same operative temperature) the decrease of the [DELTA]T between supply and return water temperatures is necessary, due to the effect of the thermal inertia in transient conditions; in this way the water energy is the same in cases a and case b: part of this energy is stored in the structure and released later in case b, while in case a the structure responds immediately. Consequently the supply water temperature for case b is lower than case a (Figure 8).
[FIGURE 8 OMITTED]
For each room hourly PMV (Predicted Mean Vote) as well as PPD (Percentage of Persons Dissatisfied) (Fanger 1970, ISO 2004) have been calculated: thermal resistance of clothing of 1 clo in heating and 0.5 clo in cooling period has been considered. In this work a residential house has been analyzed, therefore the hypothesis of a range for comfort conditions of -0.7<PMV<+0.7 has been made; only hours when persons are in the room have been considered (De Carli and Olesen 2002) according to EN 15251 (CEN 2007). During heating period, in all rooms the comfort conditions are the same for the three radiant systems. In Figure 9 the cold, hot and comfort hours in Room 1, at the second floor, are shown: the result is the same for all analyzed climates--for each hour, PMV index is in the considered comfort range..
[FIGURE 9 OMITTED]
ANNUAL COOLING ENERGY DEMAND
During the cooling period, an on/off control on the indoor temperature with set-point of 26[degrees]C (78.8[degrees]F) has been considered. As it can be seen in Table 7, the supply water temperature is almost the same in each considered city and also for each radiant system.
Table 7. Seasonal Mean Supply and Return Water Temperatures for Radiant Systems in Cooling Radiant Floor Location T input T output [[degrees]C] [[degrees]C] ([degrees]F) ([degrees]F) Wuerzburg 20.2 (68.4) 22.1 (71.8) Venice 20.2 (68.4) 21.9 (71.4) Naples 20.2 (68.4) 21.9 (71.4) Radiant Wall case a case b Location T input T output T input T output [[degrees]C] [[degrees]C] [[degrees]C] [[degrees]C] ([degrees]F) ([degrees]F) ([degrees]F) ([degrees]F) Wuerzburg 20.2 (68.4) 21.8 (71.2) 20.3 (68.5) 22.0 (71.6) Venice 20.1 (68.2) 21.7 (71.1) 20.1 (68.2) 21.8 (71.2) Naples 20.1 (68.2) 21.7 (71.1) 20.1 (68.2) 21.8 (71.2)
In Figure 10 the specific yearly cooling energy need of the building is reported (in absolute value); in Table 8 mean COP and the amount of working hours of air to water heat pump in cooling season are shown.
Table 8. Mean COP and Working Hours of Air to Water Heat Pump in Cooling Radiant Floor Radiant Wall case a case b Location COP Working hours COP Working hours COP Working hours [-] [-] [-] Wuerzburg 4.30 1368 4.17 2056 4.23 2004 Venice 4.25 2250 3.92 2976 3.92 2969 Naples 4.17 2345 4.06 3017 4.07 3016
[FIGURE 10 OMITTED]
In Figure 11 the specific annual primary energy demand for cooling can be seen (the area of bathrooms and corridors has not been considered). In Table 9 the ratio between energy removed by inside surface and overall water energy is reported. It is possible to see that the losses of radiant floor, with the same water energy, are lower than radiant wall; the difference decreases in warmer climates. The ratio greater than 1 is due to the contribution of the back side of the structure.
Table 9. Losses (1) of Radiant Systems in Cooling Mode Radiant Floor Radiant Wall Location case a case b [-] [-] [-] Wuerzburg 1.83 1.15 1.15 Venice 1.37 0.98 0.98 Naples 1.32 0.97 0.97 (1) ratio between radiant useful energy on the inside surface and water energy at pipes level.
[FIGURE 11 OMITTED]
About the relative humidity in the rooms, in Wuerzburg it is lower than 70%; in the other analyzed cities it exceeds the 70% for a few hours, when outdoor conditions are severe.
During cooling period the indoor conditions are different: in the case of radiant walls the percentage of unsatisfied persons (PPD index) is higher than radiant floor. In Figure 12 the cold, hot and comfort hours in Room 1 (Figure 1) at the first floor are shown. It is interesting to see that in Venice and Naples radiant floor ensures number of comfort hours higher than radiant wall; moreover radiant wall b improves comfort conditions when compared to the case a, even if in little amount. To improve the comfort conditions in the case of radiant walls there are two possibilities:
[FIGURE 12 OMITTED]
* supply water temperature must be lower than 20[degrees]C (68[degrees]F), but in this case there may be problems of moisture that can cause moulds on the internal surfaces if there isn't a dehumidification system: in this case part of cooling load of the rooms is moved to the air which is supplied at lower temperature;
* radiant surface must be greater, but considering that radiant wall system is put generally on the external walls, and considering also heating mode and furniture, this possibility is rather difficult to obtain
A radiant floor and two possible radiant wall systems have been compared during one year, under the same conditions, installed in a residential house in three different climates, with the purpose to evaluate the primary energy consumption with air to water heat pump. The volume of the insulation material on the building envelope is the same for the three systems and it has been conveniently distributed.
Radiant wall surface area is less than radiant floor, therefore, in comparison with radiant floor, radiant wall runs always at higher temperature in heating conditions and at lower temperature in cooling conditions. This has to be taken into account when considering distribution and production efficiencies. In fact this is evident with radiant wall case b (additional insulation layer put on the external side of the wall): the supply water temperature is lower than case a (additional insulation layer put on the inner side of the wall) and part of heat is stored into mass and released later on.
By this study it comes out that if building envelope is built with good insulation the considered radiant systems are equivalent in heating period, but it has to be underlined that additional insulation (more than recommended or calculated for heat losses as e.g. through EN 12831 (CEN 2003)) has to be put when using radiant walls.
If radiant wall system is designed for heating, the surface may be insufficient for cooling; to improve indoor comfort the increase of surface or a lower supply water temperature it is needed, but in this case there may be moisture problems. During cooling period if the additional insulation layer is put on the outside of the wall, there is a small improvement on comfort conditions.
Blomberg T. 1999. Heat2-A PC-program for heat transfer in two dimensions. Manual with brief theory and examples. Lund Group for Computational Building Physics, Sweden.
Brunello P., G. Di Gennaro, M. De Carli, and R. Zecchin. 2001. Mathematical modelling of radiant heating and cooling with massive thermal slab. Proceedings of Clima 2000, Naples.
CEN. EN 1264 - 4. 2001. Floor heating - Systems and components--Installation. Brussels: CEN European Committee for Standardization.
CEN. EN 12831. 2003. Heating systems in buildings - Method for calculation of the design heat load. Brussels: CEN European Committee for Standardization.
CEN. EN 15377 - 1. 2008. Heating systems in buildings - Design of embedded water based surface heating and cooling systems - Part 1: Determination of the design heating and cooling capacity. Brussels: CEN European Committee for Standardization.
CEN. 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. Brussels: CEN European Committee for Standardization.
De Carli M. 2002. New technologies in radiant heating and cooling. Doctoral Thesis, University of Padova.
De Carli M, Olesen BW. 2002. Long term evaluation of the general thermal comfort conditions. Proceedings of the 9th International Conference on Indoor Air Quality and climate - Indoor Air 2002, Monterey-California.
Fanger P.O. 1970. Thermal Comfort - Analysis and Applications in Environmental Engineering. Copenhagen: Danish Technical Press.
Kusuda T. 1969. Thermal response factors for multi-layer structures of various heat conduction systems. ASHRAE semi annual Meeting, Chicago.
ISO. 2004. Ergonomics of the thermal environment - Analytical determination and interpretation of thermal comfort using calculation of the PMV and PPD indices and local thermal comfort. ISO/DIS 7730. Geneve: International Organization for Standardization.
Scarpa M. 2007. Cooling of buildings by means of thermally activated building systems. Doctoral Thesis, University of Padova.
Michele De Carli, PhD
Angelo Zarrella, PhD
Michele De Carli is an assistant professor, Angelo Zarrella is a research fellow, and Roberto Zecchin is a full professor with the Dipartimento di Fisica Tecnica at the University of Padova, Padova, Italy.
|Printer friendly Cite/link Email Feedback|
|Author:||De Carli, Michele; Zarrella, Angelo; Zecchin, Roberto|
|Date:||Jul 1, 2009|
|Previous Article:||Comparative analysis of optimization approaches to design building envelope for residential buildings.|
|Next Article:||Condensate harvesting from large dedicated outside air-handling units with heat recovery.|