- Email: [email protected]
Experimental assessment of a zinc-titanium ventilated façade in a Mediterranean climate
Energy and Buildings 69 (2014) 525–534
Contents lists available at ScienceDirect
Energy and Buildings journal homepage: www.elsevier.com/locate/enbuild
Experimental assessment of a zinc-titanium ventilated fac¸ade in a Mediterranean climate Francesca Stazi a,∗ , Ambra Vegliò a , Costanzo Di Perna b a Dipartimento di Ingegneria Civile, Edile e Architettura (DICEA), Facoltà di Ingegneria, Università Politecnica delle Marche, Via Brecce Bianche, 60131 Ancona, Italy b Dipartimento di Ingegneria Industriale e Scienze Matematiche, Facoltà di Ingegneria, Università Politecnica delle Marche, Via Brecce Bianche, 60131 Ancona, Italy
a r t i c l e
i n f o
Article history: Received 27 June 2013 Received in revised form 8 November 2013 Accepted 9 November 2013 Keywords: Ventilated fac¸ades Experimental study Ventilation channel Energy saving Energy efficient building Solar passive design
a b s t r a c t An experimental study was carried out on ventilated fac¸ades with zinc-titanium cladding in a hot-summer Mediterranean climate. The aim was to investigate the thermo-physical performance, verifying the effect of the following parameters: (i) the height of the ventilation channel; (ii) the external climate conditions (direct solar radiation, wind); (iii) the exposure to sunlight; (iv) the type of external facing. The study involved the simultaneous monitoring of walls with zinc-titanium cladding with ventilation channels of different heights (4 m, 8 m and 12 m) and different exposure (south, east and west). The data obtained were also compared with those measured on ventilated walls with clay cladding (12 m south-facing wall). The results allowed to experimentally verify for the studied fac¸ades the strong relationship between difference in internal-to-external air temperature and airflow rate and to demonstrate that while the wind pressure strongly influences the lower walls airflow rate, it does not affect the higher walls performance. The exposure of the wall causes only a slight shift in the onset of the stack effect. The Reynolds number was calculated and the different air flow conditions in the ventilation channel were identified. Linear relationship between the external air temperature and air temperature in the gap were identified for each studied wall. The qualitative comparison with ventilated fac¸ades characterised by a massive clay cladding showed that the cladding’s inertia influences the time in which the stack effect becomes more effective: during the night for low inertia claddings and during the daytime for massive ones. © 2013 Elsevier B.V. All rights reserved.
1. Introduction New European Directives [1,2] on energy performance, to promote energy efficiency within the European Union, underline the problems caused by the increasing proliferation of air conditioning systems in European countries and stress the importance of adopting strategies which contribute to improving the thermal performance of buildings during the summertime. In this regard, ventilated fac¸ades (a recently developed technology for passive cooling) are of considerable interest since they help to improve the internal climate conditions and determine a reduction in the use of thermal energy during the summertime. In hot-summer Mediterranean climates, ventilated fac¸ades (especially those with opaque facing) have recently aroused great interest because they resolve the problem of the durability of the outer finishes of the external insulation layer (mainly caused by cracking as a result of aggressive
∗ Corresponding author. Tel.: +39 071 2204783; fax: +39 071 2204378; mobile: +39 328 3098217. E-mail addresses: [email protected], [email protected] (F. Stazi). 0378-7788/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.enbuild.2013.11.043
solar radiation) without the drawback of summertime overheating which is necessarily caused by double skin glazed fac¸ades. Ventilated fac¸ades with opaque cladding are generally characterised by the presence of one continuous insulation layer next to the internal mass and another external layer of protective cladding which is fastened to the wall using mechanical systems. A naturally ventilated channel is thus created between the insulation layer and the cladding. Many authors have used analytical models to describe how the thermal and energy performance of ventilated fac¸ades is influenced by some characteristics: i. the width and height of the ventilation channel [3,4]; ii. the factors connected with the site (solar radiation, wind and exposure) which determine the local microclimate [5,6]; iii. the type of external cladding, the characteristics of the materials which are laid adjacent to the channel [7,8] and the material of the inner layer [7]. The cladding may be made of a thin metal (e.g., zinc-titanium) or a thicker solid material (e.g., brick, ceramics, cement) [9] and may be permeable or watertight [10]. The behaviour of the ventilation channel may vary according
526
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
to the reciprocal positions of the thermal insulation and the inertial mass. Recent experimental studies on ventilated fac¸ades mostly describe the development of mock-ups aimed at validating a simulation model. In 2011 one group of researchers [9] developed a prototype for comparing ventilated walls with and without radiant barriers, focusing their attention on the convergence between measured and simulated values. Another group of researchers [11] presented a study of ventilated walls with metal cladding, demonstrating, by means of a scale model, the importance of climate variables (sun and wind). The work by Giancola et al. [12] is of considerable interest since it was carried out on a real case study in a Mediterranean climate. The wall analysed has a clay cladding, is south-facing and has a maximum height of 5 m. Studies on ventilated walls with integrated PV panels have recently been published [13,14]. A survey of the available literature showed that there is a lack of experimental studies on real buildings aimed at quantifying the importance of the features specified below which some authors have only investigated analytically. The current research study is aimed at experimentally assessing the efficiency of ventilated fac¸ades with metal zinc-titanium cladding, verifying the effect on performance of several different parameters: (i) the height of the ventilation channel; (ii) the factors connected with the site (exposure to sun and wind); (iii) the type of cladding. The study identifies the trends in air velocity inside the channel and the temperature distribution within the layers, for all the cases studied. These data may be useful for other authors for the calibration of numerical models of ventilated walls, which are extremely complex given the considerable variability in performance caused by the non-linearity of the processes involved. 2. Stages and methods 2.1. Stages The current experimental study, carried out during the summertime on a school building, involved the following activities: - contemporary summertime monitoring of ventilated walls with zinc-titanium cladding, characterised by different heights of the ventilation channel (4 m, 8 m and 12 m) and different exposure (south, east and west); data processing and assessment of the thermo-physical behaviour; - qualitative comparison with the data measured during a previous experimental campaign on a ventilated wall with clay cladding (12 m south-facing wall). 2.2. Case studies The building used for the case study (Fig. 1) is located in Ancona (Central Italy – latitude: 43◦ 35 03.72 N; longitude: 13◦ 31 29.79 E; altitude: 67 m), characterised by a hot-summer Mediterranean climate (Köppen climatic classification) and by 1688 degree days (climate zone D [15]). The school has a cylindrical main building with four floors above ground level used as offices and a number of lower structures with one or two floors (hmax = 8 m) which house the laboratories and classrooms. Some of the walls of the building are ventilated and present a zinc-titanium cladding. To verify the effect of the type of cladding, the data obtained during the experimental campaign were compared with the findings from a previous study [16] of ventilated walls with an external
clay fac¸ade, carried out on a school building located in the same climate area. The layers of the studied walls are shown in Fig. 2 and Table 1. 2.3. Experimental methods Monitoring was carried out during the summertime between 22 July and 29 August 2011. The following measurements (Fig. 3) were performed in accordance with UNI EN ISO 7726:2002 [17]: 1. investigation of the outdoor environmental conditions adjacent to the walls using external weather stations with direct and global pyranometers, a combined sensor for the speed and the direction of the wind and a thermo-hygrometer; 2. detailed analysis of the thermo-physical conditions of all the ventilated walls at the inlet openings and at the ventilation channel mid-height by means of: - a set of RTD sensors to measure the surface temperatures of the different layers; - hot-sphere thermo-anemometers to record the velocity and the temperature of the air in the ventilation channels; - heat flux metres to calculate the transmittances and to measure the incoming and outgoing heat flux through the insulated wall (positioned on its internal side); 3. investigation of the internal environmental conditions by means of micro-climate stations with a thermo-hygrometer to record the temperature and the relative humidity of the internal air. The characteristics of the probes are as follows: - thermo-resistances: tolerance in accordance with IEC 751; - heat flux metres: tolerance in accordance with ISO 8302; sensitivity of 50 V/(Wm2 ); - hot-sphere anemometers: accuracy of 0.03 m/s; - pyranometers: uncertainty < 2%. Data takers DT500 were used with: - voltage: resolution 1.3 V; range ± 25 mV; tolerance ± 0.16% of full scale; - RTDs, 4-Wire: resolution 0.01 ◦ C; range Pt100 (100 ); tolerance ± 0.17% of full scale; - analogue to digital conversion: accuracy 0.15% of full scale; linearity 0.005%. 3. Results and discussion 3.1. Climate conditions Fig. 4 shows the climate conditions of a significant period extrapolated from the summertime monitoring data. The global solar radiation, the external air temperature and the wind speed are reported. The period is generally characterised by sunny days with only a few days of variable or cloudy weather conditions. The external air temperature on a typical day ranges between 20 ◦ C and 30 ◦ C, with peaks of up to 37 ◦ C on some of the hottest days and minimum temperatures of 13 ◦ C during the night-time. The maximum wind speed ranges between 1 m/s and 5 m/s. 3.2. Effect of the wind and solar radiation with respect to the height of the ventilation channel Fig. 5 shows the temperature and the air velocity in the ventilation channel for the two south-facing walls of 4 m and 12 m for a selected period.
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
527
Fig. 1. View of the case study from the south.
In order to study the difference for the two walls, two sunny days were chosen (August 22 and 25) with similar global solar radiation and external air temperature but respectively characterised by the absence of wind (very low wind velocity) and the presence of wind (very high wind velocity with prevailing direction perpendicular to the studied fac¸ade). In Fig. 6 the solar irradiance on vertical surface, the external surface temperature, the temperature and the velocity of the air in the ventilation channel are reported for the 4 m and 12 m southfacing walls. The external air temperature and the wind velocity are also reported.
On both days the maximum solar irradiance measured on the vertical surface of 4 m wall is about 700 W/m2 , while on the 12 m wall is 550 W/m2 . This difference is due to the fact that the sensors positioned at the walls mid-height are at about 6 m above ground level in the higher one and at about 2 m above ground level in the lower one. As a consequence the latter is affected by a stronger groundreflected radiation. This different boundary condition affects the external surface temperature trends: the maximum temperature values measured recorded on the surface of the 4 m wall are higher than those recorded for the 12 m wall, with a maximum of 67–68 ◦ C for the former and of 54–56 ◦ C (up to 15 ◦ C lower) for the latter.
Fig. 2. Vertical section and radiation properties of the ventilated fac¸ades: the zinc-titanium (a) and the clay-cladding (b).
528
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
Table 1 Layers making up the zinc-titanium and the clay-cladding ventilated fac¸ades. Layers
Thickness [m]
Thermal conductivity [W/(m K)]
Heat capacity [J/(kg K)]
Density [kg/m3 ]
Zinc-titanium ventilated fac¸ades Gypsum plaster Hollow brick Air cavity XPS panel Mortar Semi-solid brick Ventilation channel Wooden batten Zinc-titanium
0.015 0.120 0.025 0.070 0.010 0.120 0.060 0.024 0.0008
0.400 0.387 – 0.039 1.000 0.500 – 0.140 110
26.667 3.225 – 0.557 100 4.167 – 450 7100
1000 717 – 30 1800 1167 – 450 7100
Clay-cladding ventilated fac¸ades Gypsum lime plaster Clay hollow brick Cement mortar EPS panel Ventilation channel Clay cladding
0.015 0.200 0.015 0.040 0.060 0.040
1000 840 1000 1400 – 814
1000 860 1300 15 – 792
0.400 0.230 0.500 0.040 – 0.05
Fig. 3. Plan with the monitored fac¸ades (a) and detailed vertical section (b) with the monitoring sensor positions.
The trend of air temperatures in the ventilation channel for the two walls is influenced by the different surface temperatures: in both days the lower wall presents higher values in the central hours of the day (up to 43 ◦ C) while the higher wall has a flatter trend (maximum value of 40 ◦ C). In the two days the trend of the air
temperature in the channel is very similar for the 12 m wall, while it is possible to note some difference for the 4 m wall: in this case in the windy day the air temperature in the channel after reaching the maximum value strongly decreases thanks to the rising of warm air because the wind enhances the stack effect.
Fig. 4. Weather data from the measurement campaign.
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
529
Fig. 5. Comparison between the two south-facing ventilated walls, with a ventilation channel of 4 m and 12 m.
The natural movement of air into the chimney is “called” stack effect and it is driven by buoyancy. Buoyancy is an upward force that occurs due to a difference in indoor-to-outdoor air density resulting mainly from temperature differences. The rising air in the chimney creates a draught that draws air from below. The greater the thermal difference in indoor-to-outdoor air temperature T and the height of the structure, the greater the buoyancy force, and thus the stack effect. The air velocity in the ventilation channel is expressed through the following equation [4] based on mass and energy balance:
v2 =
2Hg(e − i ) i (fH/d + 1.5)
(1)
where v = mean air velocity in the cavity [m/s]; H = height of the cavity [m]; e , i = external and internal air density [kg/m3 ] (depending on air temperature, pressure and humidity); g = gravitational acceleration [m/s2 ]; f = coefficient of friction; d = equivalent diameter [m].
To understand the behaviour of a wall it is important to analyse the temperature difference between the outside air and the air in the channel that is the driving force for the stack effect. In both walls it is possible to note from the first hours of the morning (about 7.00 a.m.) up to about mid-afternoon an increase of the values of the air temperature in the gap. There is a slight time shift between the maximum values for the two walls as a result of a small exposure difference (maximum value of 43 ◦ C reached at 1.00 p.m. for the 4 m wall and maximum values of 40 ◦ C reached at 3.00 p.m. for the 12 m wall). In this range of time the external air temperature increases significantly in the first hour (from 7.00 a.m. to 8.00 a.m.), almost equalling the temperature inside the channel. The consequent vanishing of the difference in internal-to-external air temperature causes a sudden reduction of the air flow rate, with velocity values being close to zero at the intersection of the curves. The outside temperature increases much less in the following hours (from 8.00 a.m. onwards), resulting in an increasing divergence of the air temperature curves in the channels. This difference activates the stack effect and determines the consequent sudden increase of the values of air velocity in both walls. After mid-afternoon (at
Fig. 6. Comparison between the two south-facing ventilated walls, with a ventilation channel of 4 m and 12 m, on two sunny days without wind (a) and with wind (b).
530
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
about 1.00 p.m. for the 4 m wall and at 3.00 p.m. for 12 m wall), the temperature in the channel decreases again while the values of outside air temperature remain constant until about 7.00 p.m., thus causing the reduction of the effectiveness of the stack effect. After 7.00 p.m. and during the night time the difference between the temperature in the channel and outside air temperature remains almost constant (about 5 ◦ C for lower wall and 7 ◦ C for the higher wall). In this period, the major difference in internal-to-external air temperature value in the higher walls (combined with the beneficial effect of a higher air column), activates the buoyancy-driven air movement in a most effective way. A comparison between the air velocities in the two chosen days indicates very different maximum values only for the lower wall, while only a slightly more variable trend for the higher wall. In fact, on the day with no wind (August 22) the velocity is on average 0.2 m/s for the 4 m wall and 0.6 m/s, with values going up to 0.8 m/s (17.00–24.00), for the 12 m fac¸ade. On the contrary, on a windy day (August 25), the air velocity in the ventilation channel for the 4 m wall is on average 0.4 m/s with peaks of up to 1.5 m/s (17.00). For the higher wall, the values are similar to the day without wind (0.6 m/s with peaks of up to 0.8 m/s). This comparison shows that the 12 m wall have similar air temperatures and velocity in the gap on the two days, regardless of the presence or lack of wind, while very different values of these parameters in the channel are found only for the lower wall. The difference is due to the increase of fluid dynamic resistance of the wall at the increasing of the chimney height. This result confirms through experimental data what has been obtained in an analytical way in the literature [4]. 3.3. Effect of direct solar radiation with respect to the height of the ventilation channel To verify the impact of direct solar radiation on the efficiency of the wall, two days were chosen with similar wind conditions (medium-high wind speed impacting on the wall) but characterised by the presence and the absence of direct solar radiation. In the two following graphs (Figs. 7 and 8) the sol-air temperature, the external surface temperature, the air temperature and velocity at mid-height of the ventilation channel, the air temperature in the inlet for the two chimney height are reported. The external air temperature value and wind velocity are also reported. The sun-air temperature was calculated with the following relation [18]: Tsol-air = Ta +
˛I he
(2)
where Ta = external air temperature [K]; ˛ = solar absorptance [–]; I = solar irradiance incident on the vertical surface [W/m2 ]; he = external surface heat transfer coefficient [W/(m2 K)]. The values used for solar absorptance ˛ and emissivity ε are reported in Fig. 2. For the higher wall (Fig. 7) the following observations can be drawn. On a cloudy day the external air temperature ranged between 18 ◦ C and 22 ◦ C while on a sunny day between 18 ◦ C and 31 ◦ C. On the cloudy day the sol-air temperature of the wall ranged between 18 ◦ C and 52 ◦ C with a variable trend, while on a sunny day between 18 ◦ C and 65 ◦ C. The external surface temperature of the wall was between 17 ◦ C and 35 ◦ C with a variable trend on the cloudy day and between 17 ◦ C and 53 ◦ C on the sunny day, with a difference of 18 ◦ C in the middle of the day. The air temperature in the ventilation channel is strongly influenced by the external surface temperature and ranges between 23 ◦ C and 27 ◦ C on the cloudy day and between 24 ◦ C and 38 ◦ C on
the sunny day, with a difference of as much as 11 ◦ C in the middle of the day. The air velocity in the ventilation channel has different trends in cloudy and sunny days depending on the difference in internalto-external air temperature. In the cloudy day this temperature difference T is small (about 3–4 ◦ C in the central hours and 7 ◦ C during the night) while in the sunny day it reaches higher values (up to 7 ◦ C in the central hours and about 9 ◦ C during the night). As a consequence the average air velocity in the ventilation channel is 0.45 m/s on the cloudy day and 0.6 m/s on the sunny day. The major fluctuations measured on the first case mainly depend on higher wind variability. In both cloudy and sunny days the air temperature at the inlet openings recorded values similar to the outside except in the sunny hours of the days (from 11.00 a.m. to 5.00 p.m.), period of time in which the values at the inlet increase considerably and slightly exceed the air temperature values recorded at the mid-height of the channel. For the lower wall (Fig. 8) the following observations can be drawn. On the cloudy day the sol-air temperature of the wall ranges between 18 ◦ C and 67 ◦ C, while on the sunny day between 18 ◦ C and 75 ◦ C. The external surface temperature of the wall is between 18 ◦ C and 55 ◦ C on the cloudy day and between 17 ◦ C and 67 ◦ C on the sunny day, with a difference of 12 ◦ C in the middle of the day. The air temperature in the ventilation channel is affected by the external surface temperature and ranges between 23 ◦ C and 38 ◦ C on the cloudy day and between 23 ◦ C and 40 ◦ C on the sunny day, with a difference of 2 ◦ C in the middle of the day. The air velocity in the ventilation channel has different trends in cloudy and sunny days, with greatest maximum values and more variability in the first case that mainly depends on different wind conditions. 3.4. Air flow Using the values of air velocity in the ventilation channel in two days with the same external temperature but characterised by the absence and presence of wind (respectively August 22 and 25) it was possible to determine the air flow conditions for both 4 m and 12 m wall (Fig. 9). A preliminary analysis indicates a range in which no air flow values are present. This gap represents a change in the air flow in the ventilation channel from laminar to turbulent. In the day without wind for the 4 m wall the flow remains low and constant and there is an average air flow of about 75 air changes per hour (ACH) throughout the whole day. For the 12 m wall there is an air flow more variable, with values higher than 300 ACH from about 12.00 a.m. to 3.00 a.m. of the following day (with maximum values of up to 380 ACH). In the day with high wind speed impacting on the fac¸ade the values are similar to August 22 for the higher wall (with the same maximum value but a slightly variable trend) while very different for the lower wall, reaching a better performance than the 12 m wall (with maximum values of up to 730 ACH). Moreover, for the 12 m wall the flow is nearly always turbulent while for the 4 m wall it is laminar without wind and turbulent in the windy day. 3.5. Effect of exposure To verify the effect of exposure to sunlight, the south-, east-, and west-facing walls, of respectively 12 m, 12 m and 8 m, were
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
531
Fig. 7. South-facing ventilated wall with 12 m air channel: comparison between two days, without sun (a) and with sun (b).
compared (Fig. 10) for a period including both sunny and cloudy days (from August 2 to August 5). The external air temperature, the external surface temperature, the internal surface temperature, the air velocity and temperature in the ventilation channel are reported. The selected days present similar external air temperature trends ranging between 18–20 ◦ C at night and reaching a maximum value of about 30 ◦ C in the central hours of the day. By comparing the external surface temperatures it can be noted that the minimum values are recorded in all cases at around 6.00 a.m., while the maximum values are found at different times: on the east-facing wall at 9.00 a.m., on the south-facing wall at 1.00 p.m. and on the west-facing wall at 5.00 p.m., reaching maximum values of 55 ◦ C, 50 ◦ C and 56 ◦ C, respectively. Therefore, even if the south-facing walls receive solar radiation for a longer time,
the east and west-facing walls have higher temperatures because of the low solar radiation angle in the first and last part of the day. A comparison of the internal surface temperatures reveals that in all three cases the values are between 25 ◦ C and 28 ◦ C with only a slight time lag and similar trends. The study of the air temperature values recorded inside the channel for the three walls and the difference T between these values and the external air temperature (that is the main cause of stack effect activation) explains the different behaviour of the air velocities on the three walls. The east-facing wall is affected by direct solar radiation sooner than the other walls thus undergoes an immediate rise in temperature in the gap suddenly activating the stack effect with a maximum velocity value reached at about 12.00 a.m. Subsequently the
Fig. 8. South-facing ventilated wall with 4 m air channel: comparison between two days, without sun (a) and with sun (b).
532
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
Fig. 9. Air flow: comparison between the two south-facing ventilated walls, with a ventilation channel of 4 m and 12 m, on two sunny days without sun (a) and with sun (b).
temperature in the gap quickly decreases almost equalling the outdoor air temperature value (that is still high) at 6.00 p.m., thus determining a considerable reduction in the stack effect with a consequent scarce performance during the night if compared to the other two walls. In the south-facing wall air temperature in the channel increases more gradually, and in the early morning hours it follows the trend of the outside temperature therefore with a resulting sudden decrease of the buoyancy effect. The maximum air velocity in the channel in this case is at about 4.00 p.m. with a subsequent gradual reduction. This wall preserves higher air temperature values in the channel than the other two walls throughout the whole day, allowing a better activation of the stack effect. In the west-facing wall the stack effect is activated later than the other two walls and since it has a lower air column it reaches lower maximum air velocity in the channel.
In conclusion the exposure affects the external surface temperature trends, modifying the peak time and the amplitude of the plot, and therefore having a different impact on the values of velocity in the channel: higher for the east-facing wall in the morning and higher for the south- and west-facing walls during the rest of the day, with a beneficial night time cooling effect for the latter. 3.6. Effect of the reciprocal position between insulating and massive layers The zinc-titanium wall and the clay wall present ventilation channels of the same height, 12 m, but a stratigraphy characterised by different reciprocal position between insulating and massive layers (Fig. 2): in the former the ventilated cavity is confined towards the outside by a cladding with low inertia and on the inner side by a massive layer (thermal insulation is laid behind it); in the latter the ventilated cavity is confined towards the outside by a
Fig. 10. Comparison between three walls with a high ventilation channel: south, east and west facing.
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
533
Fig. 11. The south-facing clay-cladding ventilated fac¸ade with a 12 m ventilation channel.
massive layer and on the inner side by a low inertia insulating layer. Although it is not possible to directly compare in a quantitative way the two walls since they have a different thermal resistance for the unequal insulating layer thickness, it is still possible to make some observations on how the behaviour of the two walls is influenced by the different massive layer position. A massive layer placed on the outer side is strongly influenced by the external temperature fluctuations while if it is placed inside a closed cavity it stores heat during the day and releases the heat during the night. In order to compare on a quality level the two ventilated walls, two days were selected with similar external air temperatures and solar radiation. Specifically, August 18 was chosen for the wall with zinc-titanium cladding while August 31 (2008) was chosen for a wall with an external clay cladding. For the zinc-titanium wall it is possible to refer to Fig. 7.
Fig. 11 reports the external air temperature, the solar irradiance on vertical surface, the sol-air temperature, the air temperature at mid-height and at the inlet openings of the ventilation channel, as well as the velocity of the air in the channel for the clay wall. The study of the external surface temperatures reveals that the clay-cladded wall has temperatures between 20 ◦ C and 44 ◦ C. The difference between the zinc-titanium (Fig. 7b) and clay cladding is particularly noticeable during the middle of the day when the zinctitanium, which is a metal rather than an inertial fac¸ade, is subject to a considerable overheating (up to 53 ◦ C). Moreover while in the titanium wall the external surface temperature reaches the maximum value approximately at 1.00 p.m. and then lowers gradually (both in the sunny and cloudy days), in the clay wall the external inertial cladding reaches the maximum temperature value and retain it for about 3 h (from 12.00 a.m. to 3.00 p.m.).
Fig. 12. Graph showing the air temperature in the gap as function of the external air temperature for the zinc-titanium ventilated fac¸ades.
534
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
This different behaviour is reflected on the air temperature trends in the gap. In the clay wall the difference between air temperature in the gap and outside air temperature is high throughout the day (approx. 10 ◦ C), optimally activating the stack effect in that period of time. It strongly lowers after 6.00 p.m. and for the whole night phase since the mass of the external cladding does not retain heat, thus strongly reducing the stack effect throughout this period of time. In conclusion the different relative positions between the insulation and mass in the two walls result in a more efficient stack effect activation during the late afternoon and during the night for the zinc-titanium wall while in the central hours of the day for the clay-cladded wall.
the walls with zinc-titanium cladding and during the daytime for the clay-cladded walls). Through the results it was possible to identify the trends in air velocity inside the channel and the temperature distribution within the layers for all the cases studied. Also it was possible to find linear relationship between air temperature in the channel and outdoor air temperature: these data may be useful for other authors for the calibration of numerical models for ventilated walls, which are extremely complex given the considerable variability in performance caused by the non-linearity of the processes involved. Acknowledgements
3.7. Experimental processing using regression equations Further experimental processing on the zinc-titanium walls using regression equations made it possible to determine the relationship between the outdoor air temperature values and the air temperature in the gap. The graph (Fig. 12) shows that there is a linear relation with a good reliability level for both the higher and the lower walls. For example for the southern 12 m fac¸ade the following relation was obtained: y = 0.7721x + 10.82; with R2 = 0.7006
(3)
where x = external air temperature [◦ C]; y = air temperature in the gap [◦ C]; R2 = coefficient of determination, ranges from 0 to 1 and indicates how well data points fit the line. 4. Conclusions An experimental study was carried out on zinc-titanium ventilated fac¸ades with ventilation channels of different heights and with different exposure; the data were compared with those obtained from a study on ventilated walls with a fac¸ade in clay cladding. The results allowed to: - experimentally verify the strong correlation between the temperature difference T in internal-to-external air and the stack effect in the studied ventilated facades with thin zinc-titanium coating; - prove that the lower walls have a worse effectiveness than the higher ones, both for the reduced air column and because they receive a greater share of the reflected radiation with respect to the overall surface, and then undergo a surface overheating which increases the temperature inside the channel thus reducing the difference T in internal-to-external air. - prove that the lower walls are more influenced by the presence of wind and only in case of high wind speed they exceed the performance of higher ones. - demonstrate that the exposure to the sun has an impact above all on the time when the stack effect occurs; - demonstrate that the type of external cladding and the characteristics of the materials which are laid adjacent to the channel influence the stack effect onset time (during the night time for
The authors wish to thank the Province of Ancona for having provided resources and the case study. Particular thanks go to the following engineers: Dr. Manarini, Dr. Cerasa and Dr. Ferrini. References [1] Directive 2010/31/EU of the European Parliament and of the Council of 19 May 2010 on the energy performance of buildings. [2] Directive 2012/27/EU of the European Parliament and of the Council of 25 October 2012 on energy efficiency, amending Directives 2009/125/EC and 2010/30/EU and repealing Directives 2004/8/EC and 2006/32/EC. [3] C. Balocco, A simple model to study ventilated facades energy performance, Energy and Buildings 34 (2002) 469–475. [4] P. Brunello, F. Peron, Modelli per l’analisi del comportamento fluidodinamico delle facciate ventilate, Fisica Tecnica Ambientale (1996) 313–324. [5] M.J. Suárez, C. Sanjuan, A.J. Gutiérrez, J. Pistono, E. Blanco, Energy evaluation of an horizontal open joint ventilated fac¸ade, Applied Thermal Engineering 37 (2012) 302–313. [6] C. Balocco, Aspetti energetici delle facciate a ventilazione naturale, in Facciate ventilate – Architettura, prestazioni e tecnologia, a cura di Frida Bazzocchi, Alinea editrice, 2002, pp. 269–326. [7] M. Ciampi, F. Leccese, G. Tuoni, Ventilated facades energy performance in summer cooling of buildings, Solar Energy 75 (2003) 491–502. [8] F. Patania, A. Gagliano, F. Nocera, A. Ferlito, A. Galesi, Thermofluiddynamic analysis of ventilated facades, Energy and Buildings 42 (2010) 1148–1155. [9] P. Seferis, P. Strachan, A. Dimoudi, A. Androutsopoulos, Investigation of the performance of a ventilated wall, Energy and Buildings 43 (2011) 2167–2178. [10] C. Sanjuan, M.J. Suárez, M. González, J. Pistono, E. Blanco, Energy performance of an open-joint ventilated fac¸ade compared with a conventional sealed cavity fac¸ade, Solar Energy 85 (2011) 1851–1863. [11] F. PeciLópez, R.L. Jensen, P. Heiselberg, M. Ruiz de Adana Santiago, Experimental analysis and model validation of an opaque ventilated facade, Building and Environment 56 (2012) 265–275. [12] E. Giancola, C. Sanjuan, E. Blanco, M.R. Heras, Experimental assessment and modelling of the performance of an open joint ventilated fac¸ade during actual operating conditions in Mediterranean climate, Energy and Buildings 54 (2012) 363–375. [13] O. Zogou, H. Stapountzis, Energy analysis of an improved concept of integrated PV panels in an office building in central Greece, Applied Energy 88 (2011) 853–866. [14] W. Sun, J. Ji, C. Luo, W. He, Performance of PV-Trombe wall in winter correlated with south fac¸ade design, Applied Energy 88 (2011) 224–231. [15] D.P.R. 26 agosto 1993, n. 412, Regolamento recante norme per la progettazione, l’installazione, l’esercizio e la manutenzione degli impianti termici degli edifici ai fini delcontenimento dei consumi di energia, in attuazione dell’art. 4, comma 4, della L. 9 gennaio1991, n. 10. [16] F. Stazi, F. Tomassoni, A. Vegliò, C. Di Perna, Experimental evaluation of ventilated walls with an external clay cladding, Renewable Energy 36 (2011) 3373–3385. [17] International standard UNI EN ISO 7726, Ergonomics of the Thermal Environment – Instruments for Measuring Physical Quantities, 2002. [18] ASHRAE Handbook, Volume Fundamentals, 2005.
Contents lists available at ScienceDirect
Energy and Buildings journal homepage: www.elsevier.com/locate/enbuild
Experimental assessment of a zinc-titanium ventilated fac¸ade in a Mediterranean climate Francesca Stazi a,∗ , Ambra Vegliò a , Costanzo Di Perna b a Dipartimento di Ingegneria Civile, Edile e Architettura (DICEA), Facoltà di Ingegneria, Università Politecnica delle Marche, Via Brecce Bianche, 60131 Ancona, Italy b Dipartimento di Ingegneria Industriale e Scienze Matematiche, Facoltà di Ingegneria, Università Politecnica delle Marche, Via Brecce Bianche, 60131 Ancona, Italy
a r t i c l e
i n f o
Article history: Received 27 June 2013 Received in revised form 8 November 2013 Accepted 9 November 2013 Keywords: Ventilated fac¸ades Experimental study Ventilation channel Energy saving Energy efficient building Solar passive design
a b s t r a c t An experimental study was carried out on ventilated fac¸ades with zinc-titanium cladding in a hot-summer Mediterranean climate. The aim was to investigate the thermo-physical performance, verifying the effect of the following parameters: (i) the height of the ventilation channel; (ii) the external climate conditions (direct solar radiation, wind); (iii) the exposure to sunlight; (iv) the type of external facing. The study involved the simultaneous monitoring of walls with zinc-titanium cladding with ventilation channels of different heights (4 m, 8 m and 12 m) and different exposure (south, east and west). The data obtained were also compared with those measured on ventilated walls with clay cladding (12 m south-facing wall). The results allowed to experimentally verify for the studied fac¸ades the strong relationship between difference in internal-to-external air temperature and airflow rate and to demonstrate that while the wind pressure strongly influences the lower walls airflow rate, it does not affect the higher walls performance. The exposure of the wall causes only a slight shift in the onset of the stack effect. The Reynolds number was calculated and the different air flow conditions in the ventilation channel were identified. Linear relationship between the external air temperature and air temperature in the gap were identified for each studied wall. The qualitative comparison with ventilated fac¸ades characterised by a massive clay cladding showed that the cladding’s inertia influences the time in which the stack effect becomes more effective: during the night for low inertia claddings and during the daytime for massive ones. © 2013 Elsevier B.V. All rights reserved.
1. Introduction New European Directives [1,2] on energy performance, to promote energy efficiency within the European Union, underline the problems caused by the increasing proliferation of air conditioning systems in European countries and stress the importance of adopting strategies which contribute to improving the thermal performance of buildings during the summertime. In this regard, ventilated fac¸ades (a recently developed technology for passive cooling) are of considerable interest since they help to improve the internal climate conditions and determine a reduction in the use of thermal energy during the summertime. In hot-summer Mediterranean climates, ventilated fac¸ades (especially those with opaque facing) have recently aroused great interest because they resolve the problem of the durability of the outer finishes of the external insulation layer (mainly caused by cracking as a result of aggressive
∗ Corresponding author. Tel.: +39 071 2204783; fax: +39 071 2204378; mobile: +39 328 3098217. E-mail addresses: [email protected], [email protected] (F. Stazi). 0378-7788/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.enbuild.2013.11.043
solar radiation) without the drawback of summertime overheating which is necessarily caused by double skin glazed fac¸ades. Ventilated fac¸ades with opaque cladding are generally characterised by the presence of one continuous insulation layer next to the internal mass and another external layer of protective cladding which is fastened to the wall using mechanical systems. A naturally ventilated channel is thus created between the insulation layer and the cladding. Many authors have used analytical models to describe how the thermal and energy performance of ventilated fac¸ades is influenced by some characteristics: i. the width and height of the ventilation channel [3,4]; ii. the factors connected with the site (solar radiation, wind and exposure) which determine the local microclimate [5,6]; iii. the type of external cladding, the characteristics of the materials which are laid adjacent to the channel [7,8] and the material of the inner layer [7]. The cladding may be made of a thin metal (e.g., zinc-titanium) or a thicker solid material (e.g., brick, ceramics, cement) [9] and may be permeable or watertight [10]. The behaviour of the ventilation channel may vary according
526
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
to the reciprocal positions of the thermal insulation and the inertial mass. Recent experimental studies on ventilated fac¸ades mostly describe the development of mock-ups aimed at validating a simulation model. In 2011 one group of researchers [9] developed a prototype for comparing ventilated walls with and without radiant barriers, focusing their attention on the convergence between measured and simulated values. Another group of researchers [11] presented a study of ventilated walls with metal cladding, demonstrating, by means of a scale model, the importance of climate variables (sun and wind). The work by Giancola et al. [12] is of considerable interest since it was carried out on a real case study in a Mediterranean climate. The wall analysed has a clay cladding, is south-facing and has a maximum height of 5 m. Studies on ventilated walls with integrated PV panels have recently been published [13,14]. A survey of the available literature showed that there is a lack of experimental studies on real buildings aimed at quantifying the importance of the features specified below which some authors have only investigated analytically. The current research study is aimed at experimentally assessing the efficiency of ventilated fac¸ades with metal zinc-titanium cladding, verifying the effect on performance of several different parameters: (i) the height of the ventilation channel; (ii) the factors connected with the site (exposure to sun and wind); (iii) the type of cladding. The study identifies the trends in air velocity inside the channel and the temperature distribution within the layers, for all the cases studied. These data may be useful for other authors for the calibration of numerical models of ventilated walls, which are extremely complex given the considerable variability in performance caused by the non-linearity of the processes involved. 2. Stages and methods 2.1. Stages The current experimental study, carried out during the summertime on a school building, involved the following activities: - contemporary summertime monitoring of ventilated walls with zinc-titanium cladding, characterised by different heights of the ventilation channel (4 m, 8 m and 12 m) and different exposure (south, east and west); data processing and assessment of the thermo-physical behaviour; - qualitative comparison with the data measured during a previous experimental campaign on a ventilated wall with clay cladding (12 m south-facing wall). 2.2. Case studies The building used for the case study (Fig. 1) is located in Ancona (Central Italy – latitude: 43◦ 35 03.72 N; longitude: 13◦ 31 29.79 E; altitude: 67 m), characterised by a hot-summer Mediterranean climate (Köppen climatic classification) and by 1688 degree days (climate zone D [15]). The school has a cylindrical main building with four floors above ground level used as offices and a number of lower structures with one or two floors (hmax = 8 m) which house the laboratories and classrooms. Some of the walls of the building are ventilated and present a zinc-titanium cladding. To verify the effect of the type of cladding, the data obtained during the experimental campaign were compared with the findings from a previous study [16] of ventilated walls with an external
clay fac¸ade, carried out on a school building located in the same climate area. The layers of the studied walls are shown in Fig. 2 and Table 1. 2.3. Experimental methods Monitoring was carried out during the summertime between 22 July and 29 August 2011. The following measurements (Fig. 3) were performed in accordance with UNI EN ISO 7726:2002 [17]: 1. investigation of the outdoor environmental conditions adjacent to the walls using external weather stations with direct and global pyranometers, a combined sensor for the speed and the direction of the wind and a thermo-hygrometer; 2. detailed analysis of the thermo-physical conditions of all the ventilated walls at the inlet openings and at the ventilation channel mid-height by means of: - a set of RTD sensors to measure the surface temperatures of the different layers; - hot-sphere thermo-anemometers to record the velocity and the temperature of the air in the ventilation channels; - heat flux metres to calculate the transmittances and to measure the incoming and outgoing heat flux through the insulated wall (positioned on its internal side); 3. investigation of the internal environmental conditions by means of micro-climate stations with a thermo-hygrometer to record the temperature and the relative humidity of the internal air. The characteristics of the probes are as follows: - thermo-resistances: tolerance in accordance with IEC 751; - heat flux metres: tolerance in accordance with ISO 8302; sensitivity of 50 V/(Wm2 ); - hot-sphere anemometers: accuracy of 0.03 m/s; - pyranometers: uncertainty < 2%. Data takers DT500 were used with: - voltage: resolution 1.3 V; range ± 25 mV; tolerance ± 0.16% of full scale; - RTDs, 4-Wire: resolution 0.01 ◦ C; range Pt100 (100 ); tolerance ± 0.17% of full scale; - analogue to digital conversion: accuracy 0.15% of full scale; linearity 0.005%. 3. Results and discussion 3.1. Climate conditions Fig. 4 shows the climate conditions of a significant period extrapolated from the summertime monitoring data. The global solar radiation, the external air temperature and the wind speed are reported. The period is generally characterised by sunny days with only a few days of variable or cloudy weather conditions. The external air temperature on a typical day ranges between 20 ◦ C and 30 ◦ C, with peaks of up to 37 ◦ C on some of the hottest days and minimum temperatures of 13 ◦ C during the night-time. The maximum wind speed ranges between 1 m/s and 5 m/s. 3.2. Effect of the wind and solar radiation with respect to the height of the ventilation channel Fig. 5 shows the temperature and the air velocity in the ventilation channel for the two south-facing walls of 4 m and 12 m for a selected period.
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
527
Fig. 1. View of the case study from the south.
In order to study the difference for the two walls, two sunny days were chosen (August 22 and 25) with similar global solar radiation and external air temperature but respectively characterised by the absence of wind (very low wind velocity) and the presence of wind (very high wind velocity with prevailing direction perpendicular to the studied fac¸ade). In Fig. 6 the solar irradiance on vertical surface, the external surface temperature, the temperature and the velocity of the air in the ventilation channel are reported for the 4 m and 12 m southfacing walls. The external air temperature and the wind velocity are also reported.
On both days the maximum solar irradiance measured on the vertical surface of 4 m wall is about 700 W/m2 , while on the 12 m wall is 550 W/m2 . This difference is due to the fact that the sensors positioned at the walls mid-height are at about 6 m above ground level in the higher one and at about 2 m above ground level in the lower one. As a consequence the latter is affected by a stronger groundreflected radiation. This different boundary condition affects the external surface temperature trends: the maximum temperature values measured recorded on the surface of the 4 m wall are higher than those recorded for the 12 m wall, with a maximum of 67–68 ◦ C for the former and of 54–56 ◦ C (up to 15 ◦ C lower) for the latter.
Fig. 2. Vertical section and radiation properties of the ventilated fac¸ades: the zinc-titanium (a) and the clay-cladding (b).
528
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
Table 1 Layers making up the zinc-titanium and the clay-cladding ventilated fac¸ades. Layers
Thickness [m]
Thermal conductivity [W/(m K)]
Heat capacity [J/(kg K)]
Density [kg/m3 ]
Zinc-titanium ventilated fac¸ades Gypsum plaster Hollow brick Air cavity XPS panel Mortar Semi-solid brick Ventilation channel Wooden batten Zinc-titanium
0.015 0.120 0.025 0.070 0.010 0.120 0.060 0.024 0.0008
0.400 0.387 – 0.039 1.000 0.500 – 0.140 110
26.667 3.225 – 0.557 100 4.167 – 450 7100
1000 717 – 30 1800 1167 – 450 7100
Clay-cladding ventilated fac¸ades Gypsum lime plaster Clay hollow brick Cement mortar EPS panel Ventilation channel Clay cladding
0.015 0.200 0.015 0.040 0.060 0.040
1000 840 1000 1400 – 814
1000 860 1300 15 – 792
0.400 0.230 0.500 0.040 – 0.05
Fig. 3. Plan with the monitored fac¸ades (a) and detailed vertical section (b) with the monitoring sensor positions.
The trend of air temperatures in the ventilation channel for the two walls is influenced by the different surface temperatures: in both days the lower wall presents higher values in the central hours of the day (up to 43 ◦ C) while the higher wall has a flatter trend (maximum value of 40 ◦ C). In the two days the trend of the air
temperature in the channel is very similar for the 12 m wall, while it is possible to note some difference for the 4 m wall: in this case in the windy day the air temperature in the channel after reaching the maximum value strongly decreases thanks to the rising of warm air because the wind enhances the stack effect.
Fig. 4. Weather data from the measurement campaign.
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
529
Fig. 5. Comparison between the two south-facing ventilated walls, with a ventilation channel of 4 m and 12 m.
The natural movement of air into the chimney is “called” stack effect and it is driven by buoyancy. Buoyancy is an upward force that occurs due to a difference in indoor-to-outdoor air density resulting mainly from temperature differences. The rising air in the chimney creates a draught that draws air from below. The greater the thermal difference in indoor-to-outdoor air temperature T and the height of the structure, the greater the buoyancy force, and thus the stack effect. The air velocity in the ventilation channel is expressed through the following equation [4] based on mass and energy balance:
v2 =
2Hg(e − i ) i (fH/d + 1.5)
(1)
where v = mean air velocity in the cavity [m/s]; H = height of the cavity [m]; e , i = external and internal air density [kg/m3 ] (depending on air temperature, pressure and humidity); g = gravitational acceleration [m/s2 ]; f = coefficient of friction; d = equivalent diameter [m].
To understand the behaviour of a wall it is important to analyse the temperature difference between the outside air and the air in the channel that is the driving force for the stack effect. In both walls it is possible to note from the first hours of the morning (about 7.00 a.m.) up to about mid-afternoon an increase of the values of the air temperature in the gap. There is a slight time shift between the maximum values for the two walls as a result of a small exposure difference (maximum value of 43 ◦ C reached at 1.00 p.m. for the 4 m wall and maximum values of 40 ◦ C reached at 3.00 p.m. for the 12 m wall). In this range of time the external air temperature increases significantly in the first hour (from 7.00 a.m. to 8.00 a.m.), almost equalling the temperature inside the channel. The consequent vanishing of the difference in internal-to-external air temperature causes a sudden reduction of the air flow rate, with velocity values being close to zero at the intersection of the curves. The outside temperature increases much less in the following hours (from 8.00 a.m. onwards), resulting in an increasing divergence of the air temperature curves in the channels. This difference activates the stack effect and determines the consequent sudden increase of the values of air velocity in both walls. After mid-afternoon (at
Fig. 6. Comparison between the two south-facing ventilated walls, with a ventilation channel of 4 m and 12 m, on two sunny days without wind (a) and with wind (b).
530
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
about 1.00 p.m. for the 4 m wall and at 3.00 p.m. for 12 m wall), the temperature in the channel decreases again while the values of outside air temperature remain constant until about 7.00 p.m., thus causing the reduction of the effectiveness of the stack effect. After 7.00 p.m. and during the night time the difference between the temperature in the channel and outside air temperature remains almost constant (about 5 ◦ C for lower wall and 7 ◦ C for the higher wall). In this period, the major difference in internal-to-external air temperature value in the higher walls (combined with the beneficial effect of a higher air column), activates the buoyancy-driven air movement in a most effective way. A comparison between the air velocities in the two chosen days indicates very different maximum values only for the lower wall, while only a slightly more variable trend for the higher wall. In fact, on the day with no wind (August 22) the velocity is on average 0.2 m/s for the 4 m wall and 0.6 m/s, with values going up to 0.8 m/s (17.00–24.00), for the 12 m fac¸ade. On the contrary, on a windy day (August 25), the air velocity in the ventilation channel for the 4 m wall is on average 0.4 m/s with peaks of up to 1.5 m/s (17.00). For the higher wall, the values are similar to the day without wind (0.6 m/s with peaks of up to 0.8 m/s). This comparison shows that the 12 m wall have similar air temperatures and velocity in the gap on the two days, regardless of the presence or lack of wind, while very different values of these parameters in the channel are found only for the lower wall. The difference is due to the increase of fluid dynamic resistance of the wall at the increasing of the chimney height. This result confirms through experimental data what has been obtained in an analytical way in the literature [4]. 3.3. Effect of direct solar radiation with respect to the height of the ventilation channel To verify the impact of direct solar radiation on the efficiency of the wall, two days were chosen with similar wind conditions (medium-high wind speed impacting on the wall) but characterised by the presence and the absence of direct solar radiation. In the two following graphs (Figs. 7 and 8) the sol-air temperature, the external surface temperature, the air temperature and velocity at mid-height of the ventilation channel, the air temperature in the inlet for the two chimney height are reported. The external air temperature value and wind velocity are also reported. The sun-air temperature was calculated with the following relation [18]: Tsol-air = Ta +
˛I he
(2)
where Ta = external air temperature [K]; ˛ = solar absorptance [–]; I = solar irradiance incident on the vertical surface [W/m2 ]; he = external surface heat transfer coefficient [W/(m2 K)]. The values used for solar absorptance ˛ and emissivity ε are reported in Fig. 2. For the higher wall (Fig. 7) the following observations can be drawn. On a cloudy day the external air temperature ranged between 18 ◦ C and 22 ◦ C while on a sunny day between 18 ◦ C and 31 ◦ C. On the cloudy day the sol-air temperature of the wall ranged between 18 ◦ C and 52 ◦ C with a variable trend, while on a sunny day between 18 ◦ C and 65 ◦ C. The external surface temperature of the wall was between 17 ◦ C and 35 ◦ C with a variable trend on the cloudy day and between 17 ◦ C and 53 ◦ C on the sunny day, with a difference of 18 ◦ C in the middle of the day. The air temperature in the ventilation channel is strongly influenced by the external surface temperature and ranges between 23 ◦ C and 27 ◦ C on the cloudy day and between 24 ◦ C and 38 ◦ C on
the sunny day, with a difference of as much as 11 ◦ C in the middle of the day. The air velocity in the ventilation channel has different trends in cloudy and sunny days depending on the difference in internalto-external air temperature. In the cloudy day this temperature difference T is small (about 3–4 ◦ C in the central hours and 7 ◦ C during the night) while in the sunny day it reaches higher values (up to 7 ◦ C in the central hours and about 9 ◦ C during the night). As a consequence the average air velocity in the ventilation channel is 0.45 m/s on the cloudy day and 0.6 m/s on the sunny day. The major fluctuations measured on the first case mainly depend on higher wind variability. In both cloudy and sunny days the air temperature at the inlet openings recorded values similar to the outside except in the sunny hours of the days (from 11.00 a.m. to 5.00 p.m.), period of time in which the values at the inlet increase considerably and slightly exceed the air temperature values recorded at the mid-height of the channel. For the lower wall (Fig. 8) the following observations can be drawn. On the cloudy day the sol-air temperature of the wall ranges between 18 ◦ C and 67 ◦ C, while on the sunny day between 18 ◦ C and 75 ◦ C. The external surface temperature of the wall is between 18 ◦ C and 55 ◦ C on the cloudy day and between 17 ◦ C and 67 ◦ C on the sunny day, with a difference of 12 ◦ C in the middle of the day. The air temperature in the ventilation channel is affected by the external surface temperature and ranges between 23 ◦ C and 38 ◦ C on the cloudy day and between 23 ◦ C and 40 ◦ C on the sunny day, with a difference of 2 ◦ C in the middle of the day. The air velocity in the ventilation channel has different trends in cloudy and sunny days, with greatest maximum values and more variability in the first case that mainly depends on different wind conditions. 3.4. Air flow Using the values of air velocity in the ventilation channel in two days with the same external temperature but characterised by the absence and presence of wind (respectively August 22 and 25) it was possible to determine the air flow conditions for both 4 m and 12 m wall (Fig. 9). A preliminary analysis indicates a range in which no air flow values are present. This gap represents a change in the air flow in the ventilation channel from laminar to turbulent. In the day without wind for the 4 m wall the flow remains low and constant and there is an average air flow of about 75 air changes per hour (ACH) throughout the whole day. For the 12 m wall there is an air flow more variable, with values higher than 300 ACH from about 12.00 a.m. to 3.00 a.m. of the following day (with maximum values of up to 380 ACH). In the day with high wind speed impacting on the fac¸ade the values are similar to August 22 for the higher wall (with the same maximum value but a slightly variable trend) while very different for the lower wall, reaching a better performance than the 12 m wall (with maximum values of up to 730 ACH). Moreover, for the 12 m wall the flow is nearly always turbulent while for the 4 m wall it is laminar without wind and turbulent in the windy day. 3.5. Effect of exposure To verify the effect of exposure to sunlight, the south-, east-, and west-facing walls, of respectively 12 m, 12 m and 8 m, were
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
531
Fig. 7. South-facing ventilated wall with 12 m air channel: comparison between two days, without sun (a) and with sun (b).
compared (Fig. 10) for a period including both sunny and cloudy days (from August 2 to August 5). The external air temperature, the external surface temperature, the internal surface temperature, the air velocity and temperature in the ventilation channel are reported. The selected days present similar external air temperature trends ranging between 18–20 ◦ C at night and reaching a maximum value of about 30 ◦ C in the central hours of the day. By comparing the external surface temperatures it can be noted that the minimum values are recorded in all cases at around 6.00 a.m., while the maximum values are found at different times: on the east-facing wall at 9.00 a.m., on the south-facing wall at 1.00 p.m. and on the west-facing wall at 5.00 p.m., reaching maximum values of 55 ◦ C, 50 ◦ C and 56 ◦ C, respectively. Therefore, even if the south-facing walls receive solar radiation for a longer time,
the east and west-facing walls have higher temperatures because of the low solar radiation angle in the first and last part of the day. A comparison of the internal surface temperatures reveals that in all three cases the values are between 25 ◦ C and 28 ◦ C with only a slight time lag and similar trends. The study of the air temperature values recorded inside the channel for the three walls and the difference T between these values and the external air temperature (that is the main cause of stack effect activation) explains the different behaviour of the air velocities on the three walls. The east-facing wall is affected by direct solar radiation sooner than the other walls thus undergoes an immediate rise in temperature in the gap suddenly activating the stack effect with a maximum velocity value reached at about 12.00 a.m. Subsequently the
Fig. 8. South-facing ventilated wall with 4 m air channel: comparison between two days, without sun (a) and with sun (b).
532
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
Fig. 9. Air flow: comparison between the two south-facing ventilated walls, with a ventilation channel of 4 m and 12 m, on two sunny days without sun (a) and with sun (b).
temperature in the gap quickly decreases almost equalling the outdoor air temperature value (that is still high) at 6.00 p.m., thus determining a considerable reduction in the stack effect with a consequent scarce performance during the night if compared to the other two walls. In the south-facing wall air temperature in the channel increases more gradually, and in the early morning hours it follows the trend of the outside temperature therefore with a resulting sudden decrease of the buoyancy effect. The maximum air velocity in the channel in this case is at about 4.00 p.m. with a subsequent gradual reduction. This wall preserves higher air temperature values in the channel than the other two walls throughout the whole day, allowing a better activation of the stack effect. In the west-facing wall the stack effect is activated later than the other two walls and since it has a lower air column it reaches lower maximum air velocity in the channel.
In conclusion the exposure affects the external surface temperature trends, modifying the peak time and the amplitude of the plot, and therefore having a different impact on the values of velocity in the channel: higher for the east-facing wall in the morning and higher for the south- and west-facing walls during the rest of the day, with a beneficial night time cooling effect for the latter. 3.6. Effect of the reciprocal position between insulating and massive layers The zinc-titanium wall and the clay wall present ventilation channels of the same height, 12 m, but a stratigraphy characterised by different reciprocal position between insulating and massive layers (Fig. 2): in the former the ventilated cavity is confined towards the outside by a cladding with low inertia and on the inner side by a massive layer (thermal insulation is laid behind it); in the latter the ventilated cavity is confined towards the outside by a
Fig. 10. Comparison between three walls with a high ventilation channel: south, east and west facing.
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
533
Fig. 11. The south-facing clay-cladding ventilated fac¸ade with a 12 m ventilation channel.
massive layer and on the inner side by a low inertia insulating layer. Although it is not possible to directly compare in a quantitative way the two walls since they have a different thermal resistance for the unequal insulating layer thickness, it is still possible to make some observations on how the behaviour of the two walls is influenced by the different massive layer position. A massive layer placed on the outer side is strongly influenced by the external temperature fluctuations while if it is placed inside a closed cavity it stores heat during the day and releases the heat during the night. In order to compare on a quality level the two ventilated walls, two days were selected with similar external air temperatures and solar radiation. Specifically, August 18 was chosen for the wall with zinc-titanium cladding while August 31 (2008) was chosen for a wall with an external clay cladding. For the zinc-titanium wall it is possible to refer to Fig. 7.
Fig. 11 reports the external air temperature, the solar irradiance on vertical surface, the sol-air temperature, the air temperature at mid-height and at the inlet openings of the ventilation channel, as well as the velocity of the air in the channel for the clay wall. The study of the external surface temperatures reveals that the clay-cladded wall has temperatures between 20 ◦ C and 44 ◦ C. The difference between the zinc-titanium (Fig. 7b) and clay cladding is particularly noticeable during the middle of the day when the zinctitanium, which is a metal rather than an inertial fac¸ade, is subject to a considerable overheating (up to 53 ◦ C). Moreover while in the titanium wall the external surface temperature reaches the maximum value approximately at 1.00 p.m. and then lowers gradually (both in the sunny and cloudy days), in the clay wall the external inertial cladding reaches the maximum temperature value and retain it for about 3 h (from 12.00 a.m. to 3.00 p.m.).
Fig. 12. Graph showing the air temperature in the gap as function of the external air temperature for the zinc-titanium ventilated fac¸ades.
534
F. Stazi et al. / Energy and Buildings 69 (2014) 525–534
This different behaviour is reflected on the air temperature trends in the gap. In the clay wall the difference between air temperature in the gap and outside air temperature is high throughout the day (approx. 10 ◦ C), optimally activating the stack effect in that period of time. It strongly lowers after 6.00 p.m. and for the whole night phase since the mass of the external cladding does not retain heat, thus strongly reducing the stack effect throughout this period of time. In conclusion the different relative positions between the insulation and mass in the two walls result in a more efficient stack effect activation during the late afternoon and during the night for the zinc-titanium wall while in the central hours of the day for the clay-cladded wall.
the walls with zinc-titanium cladding and during the daytime for the clay-cladded walls). Through the results it was possible to identify the trends in air velocity inside the channel and the temperature distribution within the layers for all the cases studied. Also it was possible to find linear relationship between air temperature in the channel and outdoor air temperature: these data may be useful for other authors for the calibration of numerical models for ventilated walls, which are extremely complex given the considerable variability in performance caused by the non-linearity of the processes involved. Acknowledgements
3.7. Experimental processing using regression equations Further experimental processing on the zinc-titanium walls using regression equations made it possible to determine the relationship between the outdoor air temperature values and the air temperature in the gap. The graph (Fig. 12) shows that there is a linear relation with a good reliability level for both the higher and the lower walls. For example for the southern 12 m fac¸ade the following relation was obtained: y = 0.7721x + 10.82; with R2 = 0.7006
(3)
where x = external air temperature [◦ C]; y = air temperature in the gap [◦ C]; R2 = coefficient of determination, ranges from 0 to 1 and indicates how well data points fit the line. 4. Conclusions An experimental study was carried out on zinc-titanium ventilated fac¸ades with ventilation channels of different heights and with different exposure; the data were compared with those obtained from a study on ventilated walls with a fac¸ade in clay cladding. The results allowed to: - experimentally verify the strong correlation between the temperature difference T in internal-to-external air and the stack effect in the studied ventilated facades with thin zinc-titanium coating; - prove that the lower walls have a worse effectiveness than the higher ones, both for the reduced air column and because they receive a greater share of the reflected radiation with respect to the overall surface, and then undergo a surface overheating which increases the temperature inside the channel thus reducing the difference T in internal-to-external air. - prove that the lower walls are more influenced by the presence of wind and only in case of high wind speed they exceed the performance of higher ones. - demonstrate that the exposure to the sun has an impact above all on the time when the stack effect occurs; - demonstrate that the type of external cladding and the characteristics of the materials which are laid adjacent to the channel influence the stack effect onset time (during the night time for
The authors wish to thank the Province of Ancona for having provided resources and the case study. Particular thanks go to the following engineers: Dr. Manarini, Dr. Cerasa and Dr. Ferrini. References [1] Directive 2010/31/EU of the European Parliament and of the Council of 19 May 2010 on the energy performance of buildings. [2] Directive 2012/27/EU of the European Parliament and of the Council of 25 October 2012 on energy efficiency, amending Directives 2009/125/EC and 2010/30/EU and repealing Directives 2004/8/EC and 2006/32/EC. [3] C. Balocco, A simple model to study ventilated facades energy performance, Energy and Buildings 34 (2002) 469–475. [4] P. Brunello, F. Peron, Modelli per l’analisi del comportamento fluidodinamico delle facciate ventilate, Fisica Tecnica Ambientale (1996) 313–324. [5] M.J. Suárez, C. Sanjuan, A.J. Gutiérrez, J. Pistono, E. Blanco, Energy evaluation of an horizontal open joint ventilated fac¸ade, Applied Thermal Engineering 37 (2012) 302–313. [6] C. Balocco, Aspetti energetici delle facciate a ventilazione naturale, in Facciate ventilate – Architettura, prestazioni e tecnologia, a cura di Frida Bazzocchi, Alinea editrice, 2002, pp. 269–326. [7] M. Ciampi, F. Leccese, G. Tuoni, Ventilated facades energy performance in summer cooling of buildings, Solar Energy 75 (2003) 491–502. [8] F. Patania, A. Gagliano, F. Nocera, A. Ferlito, A. Galesi, Thermofluiddynamic analysis of ventilated facades, Energy and Buildings 42 (2010) 1148–1155. [9] P. Seferis, P. Strachan, A. Dimoudi, A. Androutsopoulos, Investigation of the performance of a ventilated wall, Energy and Buildings 43 (2011) 2167–2178. [10] C. Sanjuan, M.J. Suárez, M. González, J. Pistono, E. Blanco, Energy performance of an open-joint ventilated fac¸ade compared with a conventional sealed cavity fac¸ade, Solar Energy 85 (2011) 1851–1863. [11] F. PeciLópez, R.L. Jensen, P. Heiselberg, M. Ruiz de Adana Santiago, Experimental analysis and model validation of an opaque ventilated facade, Building and Environment 56 (2012) 265–275. [12] E. Giancola, C. Sanjuan, E. Blanco, M.R. Heras, Experimental assessment and modelling of the performance of an open joint ventilated fac¸ade during actual operating conditions in Mediterranean climate, Energy and Buildings 54 (2012) 363–375. [13] O. Zogou, H. Stapountzis, Energy analysis of an improved concept of integrated PV panels in an office building in central Greece, Applied Energy 88 (2011) 853–866. [14] W. Sun, J. Ji, C. Luo, W. He, Performance of PV-Trombe wall in winter correlated with south fac¸ade design, Applied Energy 88 (2011) 224–231. [15] D.P.R. 26 agosto 1993, n. 412, Regolamento recante norme per la progettazione, l’installazione, l’esercizio e la manutenzione degli impianti termici degli edifici ai fini delcontenimento dei consumi di energia, in attuazione dell’art. 4, comma 4, della L. 9 gennaio1991, n. 10. [16] F. Stazi, F. Tomassoni, A. Vegliò, C. Di Perna, Experimental evaluation of ventilated walls with an external clay cladding, Renewable Energy 36 (2011) 3373–3385. [17] International standard UNI EN ISO 7726, Ergonomics of the Thermal Environment – Instruments for Measuring Physical Quantities, 2002. [18] ASHRAE Handbook, Volume Fundamentals, 2005.