retrofitted building such as a This work has been referred to advantage the change active and protection. protection and night-time

Transcription

1 Development of an innovative retrofitted building window for Catherine Guigou-Carter, François Olive, Michel Perraudeau, Roland Wetta CSTB, 24 Rue Joseph Fourier, Saint Martin d Hères, France. Fabio Sicurella INES RDI LOCIE CNRS FRE 3220, Bâtiment LYNX, Parc technologique de Savoie Technolac, 50 Avenue Lac Leman, BP 33, Le Bourget du Lac Cedex, France. Summary The aim of the Actifen-Rehab project has been to define, to implement and to test in-situ an innovative window for retrofitted building. This innovative window includes a day-time solar protection system associated to a night-time high airflow rate ventilation system; this windoww brings innovative solutions with respect to ventilation, thermal, acoustic and visual comfort. The performance of a demonstration version of this innovative window has been evaluated on two experimental sites. The window ested is based a low emissivity 4/16/4 double pane window. A 10% perforated clear membrane was used for solar protection. The ventilation system was set to achieve a m 3 /h airflow rate. In order to achieve acoustic French regulation in terms of servicee equipment noise in the receiving room, a single channel active sound control system was implemented and experimentally tested on high airflow rate ventilation system prototype to reducee the low frequency noise level mostly associated with the fan blade passage frequency. A regulation system integrating a coherent regulation scenario for the day-time solar protection system and the night-time high airflow rate ventilation system was designed. The complementarities of the night- time high airflow rate ventilationn system and the day-time solar protection system on the interiorr temperature variation were confirmed especially for hot sunny days. The acoustic performance of the façade equipped with the Actifen-Rehab window is very close to the one equipped with a standard window. Thus, the relevance of this new innovative window with the developed regulation system has been demonstrated. PACS no Jh, Ki 1. Introduction This work has been carried out within the ACTIFEN-Réhab project aiming to the development of an active window for building rehabilitation. In the context of an existing building, it is of interest to take advantage of the change of an old window, to install an active and smart window bringing innovative solutions with respect to ventilation, thermal, acoustic and visual comfort. In order to limit the use of air-conditioning systems that are quite energy-consuming, a system including a solar protection and a ventilationn apparatus was investigated in this project. The solar protection is to be used during daylight, limiting temperature increase associated with the sun in the building room where the active window is mounted. This solar protection also allows controlling the glare associated with sun light in the room. A high airflow rate ventilationn system is used during night-time in order to cool down a building room such as a bedroom during summer night without actually opening the window. These two elements, solar protection and high airflow rate ventilation, are to be includedd in the active window, referred to as the ACTIFEN-window, and automatically regulated (except in some specific cases). The performance of the ACTIFEN-window has been evaluated on two experimental sites (Nantes and Chambery in France). On both sites, two rooms were evaluated: one room was equipped with the new ACTIFEN-window and the other room with a similar standard window. The ACTIFEN-windoww tested on both sites is based a low emissivity 4/16/44 double pane window (1 m² window panes), whichh acoustic performance corresponds to an index R A,,tr 28 db (without air-inlet), and the thermal performance corresponds to the required thermal transmission coefficient (U=1.1 W/m 2 K). A 10% perforated clear membrane was used as solar protection. (c) European Acoustics Association, ISBN: , ISSN:

2 Some aeraulic measurements allowed determining the ventilation fan supply voltage to achieve an airflow rate in the order of m 3 /h. In order to achieve acoustic French regulation in terms of service equipment noise corresponding to a level L nat of 30 db(a) in the receiving room, a single channel active sound control system was implemented and experimentally tested on high airflow rate ventilation system prototype to reduce the low frequency noise level mostly associated with the fan blade passage frequency. A regulation system integrating a coherent regulation scenario for the day-time solar protection system and the night-time high airflow rate ventilation system was designed, conceived and realized in order to have the ACTIFEN-window prototype working as specified. The first section of this paper presents the high airflow rate ventilation system developed in the project and its corresponding acoustic performance. The second section discusses the in-situ evaluation of the ACTIFEN-window. 2. High airflow rate ventilation system To be installed in a building, the high airflow rate ventilation system has to comply with the current French acoustic regulation: the service equipment noise level global index L nat should not exceed 30 db(a). A schematic of the high airflow rate ventilation system within the ACTIFEN-window is shown in Figure 1. It is composed of an L-shaped duct system located on the outside; a fan is placed at one end and an air outlet centered in the horizontal lower frame part allows the air flow to enter the room to be cooled down. This high airflow rate ventilation system is to be integrated in the window frame. A radial fan has been chosen since it can be more easily integrated in the duct system and was found to yield better performances in terms of airflow rate, radiated noise and power supply. The fan voltage level is chosen to achieve the specified airflow rate of 130 m 3 /h; aeraulic measurements were performed to evaluate the effect of different treatments included in the system on the airflow rate. Passive materials are indeed used in order to decrease the mid and high frequency noise level. A diagonal filter was introduced in the duct in order to filter the air particles entering the duct; absorbing material (glass wool type) was also placed along the duct sides as shown in Figure 2. A single channel active noise control system was implemented to reduce the low frequency noise level mostly associated with the fan blade passage frequency. Active Noise Control (ANC) [1] has been applied successfully in the past to onedimensional systems such as ventilation ducts [2] and industrial chimneys [3]. Previous work at CSTB has been involved with the development of natural ventilation systems consisting of rectangular duct openings equipped with passive and active noise control technologies, in order to maintain sufficient insulation levels for buildings in the vicinity of airports, and air flow levels comparable of those of an open window [4-5]. Figure 1. Schematic of the high airflow rate ventilation system in ACTIFEN-window. Diagonal filter Radial fan Absorbing material inside duct along vertical part below diagonal filter Duct Air outlet Radial Fan (noise source) Reference microphone Error microphone Air outlet Loudspeaker (control source) Figure 2. Passive elements placed in duct ventilation system along with the active control components. The active control system includes a speaker as a secondary source; it is mounted with its enclosure on the outside of the duct on the horizontal part 1836 (c) European Acoustics Association, ISBN: , ISSN:

3 close to the bend as shown in Figure 2. The error microphone is positioned in the duct centered with respect to the air outlet aperture. The controller used for the experiments is a single channel feedforward controller based on the Filtered-X LMS algorithm [1]. This time domain controller requires a reference signal coherent with the disturbance source. This signal is provided by an additional microphone positioned below the fan inside the duct. This reference microphone cannot be placed too close to the fan due to aerodynamic turbulences (inducing coherence problem for the control) but should be sufficiently far away from the control plane in order to ensure a causal control filter. The optimal location was found to be in the middle of the duct section about 30 cm below the fan. Such a duct system mounted with the fan and the air outlet going into a reverberant room through a wall has been tested in a laboratory. The average acoustic pressure level is obtained in the reverberant room with a microphone placed on a rotating microphone boom (spatio-temporal average). The sound pressure level associated to the equipment, L nt in db, is obtained from the expression L nt = L averaged pressure -10 Log 10 (Tr/0.5), (1) i.e. the averaged pressure level with a normalized reverberation time of 0.5s. The associated global index is denoted L nat and is expressed in db(a) (A-weighting corrections between the third octave bands 100 and 5000 Hz). With respect to the French acoustic regulation, the global index L nat should be lower or equal to 30 db(a). Figure 3 presents the active control effect. A highpass filter is used on the error microphone to limit the low frequency impact (below 50 Hz) on the active control convergence. The maximum of sound pressure level in third octave band 250 Hz is lowered by more than 10 db; the decrease of sound pressure level is more than 9 db from the third octave band 200 to 400 Hz and around 4 db between 500 and 630 Hz. Thus, the global index L nat is reduced from 36 db(a) before active control is applied to 29 db(a) when active control is implemented, yielding a 7 db(a) improvement. Therefore, the goal in terms of equipment noise corresponding to French regulation requirement, i.e. L nat 30 db(a), is achieved. 3. In-situ evaluation of ACTIFENwindow In this section, the in-situ evaluation of the ACTIFEN-window performed on the Chambery site is presented. Two separate rooms were evaluated: one room was equipped with the ACTIFEN-window (see Figure 4) and the other room with a similar standard window. These two separated rooms, with a façade facing south, are adiabatic, so they are perfectly insulated; no thermal energy exchange is possible except through the window in the façade. The tested ACTIFEN-window is based a low emissivity 4/16/4 double pane window (1 m² window panes), which acoustic performance corresponds to an index R A,tr 28 db without air-inlet, and the thermal performance corresponds to the required thermal transmission coefficient (U=1.1 W/m 2 K). A 10% perforated clear membrane was used as solar protection and placed on the window outside. It was however not possible at that stage to include the ventilation system in the window frame; therefore it was directly mounted in the experimental room façade (see Figure 4). The measurements concerned with thermal, light and acoustic performance were completed during the 2010 summer. LnT (db) Frequency (Hz) Before control - LnAT=36 db(a) With control - LnAT=29 db(a) Figure 3. Sound pressure level in third octave band - Effect of active control The developed regulation system to pilot the ACTIFEN-window integrates a coherent regulation scenario for the day-time solar protection system and the night-time high airflow rate ventilation system. It is based on information from several sensors: A day/night detector for the ventilation system which is turned on only at night Indoor and outdoor temperature transducers for the ventilation system which is turned on only if the indoor-outdoor temperature difference is above a defined level Three pyranometers to evaluate the horizontal diffuse sunlighting radiation, the horizontal global sunlighting radiation, and the vertical (c) European Acoustics Association, ISBN: , ISSN:

4 global sunlighting radiation, in order to move up and down the solar protection Figure 4. Façade equipped with ACTIFEN-window on Chambery experimental site. The month of July was used to validate the different regulation mechanisms, as well as the choice of different parameters. The month of August was dedicated to the performance measurements. During that month, the ventilation system was turned on 22.5% of the time, while the solar protection was down 13.7% of the time Room temperature improvement Table I presents the time percentages during which the indoor temperature of the two rooms for the month of August was below 22 C, between 22 C and 27 C (temperature comfort zone), and above 27 C (temperature discomfort zone because of the heat). In order to evaluate the system potential for different applications (offices and apartments for example), these percentages have been calculated for different time intervals: 24h interval, from 8am to 6pm (day-time), and from 6pm to 8am (nighttime). In all cases (with small difference between day-time and night-time use), the ACTIFENwindow regulation system allows obtaining an important improvement of the thermal comfort: the time percentage during which the temperatures are in the comfort zone (between 22 and 27 C) increases from 54.6% (for the standard window) to 81.3% (for active window). All uncomfortable situations associated to heat (temperatures above 27 C) disappear with the use of ACTIFENwindow. The complementarities of the night-time high airflow rate ventilation system and the day-time solar protection system on the interior temperature variation have also been demonstrated especially for hot sunny days Indoor lighting performance The use of the solar protection reduces by 70-80% the indoor lighting and limits the lighting values to about 300 lux. The use of interior artificial lighting depending on the usage of the room equipped with the ACTIFEN- window will thus probably be necessary. Indeed, for a large time percentage (more than 90%), the level of artificial lighting (100% and even 50%) is not reached when the solar protection covers the window. However, the automatic functioning of the ACTIFEN- window during day-time will most likely take place when nobody is in the room equipped with this active window; so the decrease in interior lighting should not be such an important problem. Table I. Comfort/Discomfort time percentages during August h time interval Room equipped with Standard window ACTIFENwindow T < 22 C 12.1% 16.1% 22 C T < 27 C 54.6% 81.3% 27 C T 33.3% 2.6% 8am to 6pm time interval (day-time) Room equipped with Standard window ACTIFENwindow T < 22 C 12.5% 17.4% 22 C T < 27 C 51.5% 79.8% 27 C T 36.0% 2.7% 6pm to 8am time interval (night-time) Room equipped with Standard window ACTIFENwindow T < 22 C 11.8% 15.2% 22 C T < 27 C 56.7% 82.3% 27 C T 31.5% 2.5% 3.3. In-situ acoustic performance The façade implemented on the experimental cells was not acoustically very performing (they were rather designed for thermal concern); it was a lightweight wood frame façade 10 cm in thickness, with a single 10 mm thick OSB board on each side of the wood frame and glass wool layer for thermal insulation. It should be noted that for thermal concern in the summer 10 cm of insulating material is sufficient. Façade sound insulation for both cells as well as the outdoor environmental noise associated with the ventilation fan were measured (c) European Acoustics Association, ISBN: , ISSN:

5 Figure 5 presents the façade sound insulation for the room equipped with a standard window and for the room equipped with ACTIFEN-window. Both results are very close and indicate that the ACTFEN-window acoustic performance is of the same order as the standard window. However, the effect of the high airflow rate ventilation system is clearly noticeable on the façade sound insulation between the third octave bands 250 and 500 Hz, which corresponds to expectations (air leakage through the ventilation system). The global index for façade sound insulation satisfies the French regulation (D nt,a,tr 30 db) : For the façade equipped with the ACTIFENwindow, D nt,a,tr = 30 db For the façade equipped with the standard window, D nt,a,tr = 31 db level L eq is increased to 47 db(a) (with no plane flying above the zone). To evaluate the disturbance at a possible neighbor, the averaged L eq level was measured at a distance of about 5 m from the ventilation fan and in front of the façade equipped with the standard window. An averaged L eq level of 42 db(a) was measured (i.e. 5 db(a) below the L eq measured in front of the ventilation fan). This corresponds to only 4 db(a) above the outdoor background noise level. Therefore, the emergence limit value of 3 db(a) above the background noise defined in the French regulation for night-time (between 10pm to 7am) will likely not be respected. Planes flying over 60 Façade equipped with standard window 50 Façade equipped with ACTIFEN-window L eq db(a) 40 DnT (db) L eq =47 db(a) with ventilation fan turned on L eq =38 db(a) outdoor background noise (with ventilation fan turned off) Time Figure 6. Averaged L eq measured outdoor in front of the façade equipped with the ACTIFEN-window Frequency (Hz) Figure 5. Façade sound insulation measured on Chambery experimental site. Figure 6 presents the averaged L eq level in db(a), when the outdoor microphone is placed in front of the ventilation fan at 2 m of the façade equipped with the ACTIFEN-window. The outdoor environmental background noise level correspond to L eq =38 db(a) when the ventilation system is turned off and when there is no plane flying above the zone (a small airport is located very close to the experimental site). A similar level is to be expected during night-time (the site being relatively quiet except for the planes not allowed for taking off and landing at night). When the ventilation system is turned on, the averaged noise Figure 7 presents the sound pressure level in octave bands for the microphone positioned at 2 m in front the ventilation fan. For the octave bands of 125 Hz and 250 Hz, the pressure level with the ventilation fan turned on, is lower than 7 db (emergence limit fixed in the French acoustic regulation for these octave bands) above the outdoor environmental background noise. For the octave bands from 500 to 4000 Hz, the pressure level with the ventilation fan turned on, should be less than 5 db (emergence limit fixed in the French acoustic regulation for these octave bands) above the outdoor environmental background noise, at approximately 4 to 5 m from the ventilation fan. An acoustic hood could be use on the ventilation fan in order to limit the disturbance and discomfort that could be associated with the fan noise in a relatively quiet area. (c) European Acoustics Association, ISBN: , ISSN:

6 4. Conclusions In this paper, the development of an active smart window for building rehabilitation was discussed. The investigated ACTIFEN-window includes a solar protection and a ventilation system. The solar protection is used during daylight, limiting temperature increase associated with the sun in a room where the active window is mounted. A high airflow rate ventilation system is used during night-time in order to cool down a room such as a bedroom during summer night without opening the window. A regulation system integrating a coherent regulation scenario for the day-time solar protection system and the night-time high airflow rate ventilation system was designed. The performance of the ACTIFEN-window was evaluated on two different experimental sites in France. In order to achieve acoustic French regulation in terms of service equipment noise corresponding to a level L nat of 30 db(a) in the receiving room, passive absorbing elements as well as active noise control were successfully implemented in the ventilation duct system. In-situ measurements during 2010 summer have demonstrated that the use of the ACTIFEN-window allows obtaining an important improvement of the thermal comfort: The complementary aspects of the night-time high airflow rate ventilation system and the day-time solar protection system on the interior temperature variation were demonstrated especially for hot sunny days. The acoustic performance of the façade equipped with the ACTIFEN-window is very close to the one equipped with a standard window. In the industrialized version of the ACTIFEN-window, the integration of the ventilation fan including an acoustic hood should easily allow that in a quite quiet area, the noise associated to the active window fan will not be perturbing. Thus, the relevance of this new ACTIFENwindow with the developed regulation system has been demonstrated. The full integration of the different elements (day-time solar protection and night-time high airflow rate ventilation system with the active control scheme) to achieve an industrialised product remains to be completed. as well as the ADEME and ANR agencies for their financial support. Sound pressure Level (db) Outdoor environmental background noise With ventilation fan turned on (microphone at 2 m) Figure 7. Sound pressure level measured at 2 m from ventilation fan in octave bands. References [1] P.A. Nelson, S.J. Elliott: Active Control of Sound. Academic Press Limited, London, [2] Ch. Carme, P. De Man, V. Delemotte: ACTA TM : The second generation of active ventilation muffler. Proc. Noise-Con 97. [3] A. L Espérance, M. Bouchard, B. Paillard, C. Guigou,,A. Boudreau: Active noise control in large circular duct using an error sensors plane. Applied Acoustics, 57(4), (1999). [4] J. Maillard, C. Guigou-Carter: Study of passive/active control on openings for natural ventilation in buildings. Proc. InterNoise [5] J. Maillard, C. Guigou-Carter, P. Ducruet, C. Martin: Hybrid window for natural ventilation in building near airports. Proc. EuroNoise Frequency (Hz) Acknowledgement The authors acknowledge gratefully all the project partners, i.e. ALDES, CETIAT, CSTB, INES- LOCIE, LAPEYERE, Photowatt and Schneider Electric, involved in the ACTIFEN-Réhab project, 1840 (c) European Acoustics Association, ISBN: , ISSN: