THE EFFECT OF OPENING DISTRIBUTION AREA MODIFICATION ON HORIZONTAL LIGHT PIPE DAYLIGHT PERFORMANCE

A deep-plan office building design limits daylight access on the workspace distant from the side window. Horizontal Light Pipe (HLP) is one of the light transport systems that can deliver daylight to these areas. The research aim was to explain and evaluate the effect of HLP’s opening distribution area on daylight performance at deep plan-private office space. The research method was experimental with simulation as a tool. Daylight level and distribution of the base case, HLP with an opening distribution area of 6.6 m2 were compared with the case, HLP with an opening distribution area of 3.41 m2. The results showed that both cases distributed daylight uniformly. A 50% reduction of HLP’s opening distribution area, from 6.6 m to 3.41 m improved the average Daylight Factor as big as 6.42%. HLP with a smaller opening distribution area can be applied as the main source of daylight on deep-plan office spaces.


INTRODUCTION
The use of daylighting within an office building in the tropics provides energy savings, physiological and psychological advantages for building occupants. Proper utilization of daylighting can reduce energy for electric lights in a typical office building (Lechner, 2015) and also cooling energy consumption (Alrubaih et al., 2013). Daylight provides vitamin D and a wellbalanced circadian rhythm (Boubekri, 2008). People also desire good daylighting in their living and working environments (Li and Lam, 2003). Reduction of absenteeism, increase productivity and financial savings are other benefits gained in daylit and full spectrum office building (Edwards &Torcellini, 2002).
According to Heerwagen (2004), in order to facilitate the admission of daylight, the building should have a narrow depth. Extending the perimeter form of a building may improve the building's performance by increasing the total daylighting area (Ander, 2003). However, plans of modern airconditioned buildings tend to be deep (Lomas, 2007) in order to minimize heat gain from building envelope. Deep-plan office building design limits access to daylighting and generates an insufficient daylight level on the workspace in areas distant from the side window.
Horizontal Light Pipe (HLP) is one of the light transport system (Kischkoweit-Lopin, 2002) and is designed to supplement the daylight admitted by a lower vision window and to be the main source of daylight at 4.6-9.1 m from side lighting (Beltran et al., 1997). Aperture, pipes, and distribution opening are the main elements of HLP. Aperture collects, redirects, or concentrates incident sunlight. Pipes transport the daylight inwards the building, while distribution opening distributes daylight into deep areas of the room (Canziani et al., 2004).
Previous research focused on daylighting performance of HLP had been conducted, such as four types of HLP in 9 m office space (Beltran et al., 1997); flat aperture HLP (Canziani et al., 2004); HLP with tiltable mirror (Hien et al, 2007) and HLP with Laser Cut Panel (Garcia Hansen et al., 2001;Kwok & Chung (2008). The combination of HLP with louver (Elsiana et al., 2015a) and HLP with branching opening distribution (Elsiana et al., 2015b) was also studied. Those research showed the ability of HLP in illuminating space distances from side lighting. Different from previous research, a single HLP was applied at a deeper office space (10.5 m). Without any access to side lighting, HLP in this research acted as the main source of daylight. The research aim was to evaluate and explain the effect of HLP's opening distribution area on daylighting performance at deep plan-private office space in the tropics.

Horizontal Light Pipe in Private Office Space
HLP type C prototype by Beltran et al., 1997 was used in this study with several improvements and different applications. A single Horizontal Light Pipe (HLP) was installed in a deeper room depth (10.5 m), consist of two identical private office spaces. Those private office spaces were located in a tropical area, Surabaya (latitude 7º15' South and longitude 112º44'33'' East), under an overcast sky condition.
Placed in a tropical area, HLP's aperture faced West to utilize daylight in the afternoon. This HLP's aperture orientation was in line with the previous research by Chirarattananon et al., 2000, and the nature of the sun path along the tropical area of Surabaya. Figure 1 shows the sun path diagram of Surabaya (stereographic diagram), which was calculated using Ecotect analysis.
Located at the distance of 4.5 m from the perimeter window, both spaces didn't have access to side lighting and depended only on HLP as the main source of daylight. Each room had 3 m in width, 4 m in length, and 2.75 m in height, as described in Figure  2 and Figure 3. Those private office spaces were placed at the center of an office building that had 24 m in width and length. Ceiling's reflectance was 85%, while the wall and floor reflectance were 70% and 40%, respectively (Rea in Egan and Olgyay, 2002). This office building was free from shadow casting from adjacent buildings and vegetation. The HLP had a trapezoidal section both in plan and elevation. The width of HLP's aperture and the rear of HLP was 1.8 m and 0.9 m, respectively. The length of the HLP was 10.5 m (Figure 3). HLP's aperture had 1.8 m in width and 0.6 m in height. The aperture was covered by a single clear glass that had Visible Transmittance (VT) of 88%. In order to redirect incoming sunlight, the aperture was equipped with central and side reflectors which had a reflectivity of 88%.
The rectilinear pipe that transports the daylight had 0.6 m in height and was covered by 95% specular reflective film on its surface. The material of opening distribution, a daylight diffuser into a deep area of the building, was a translucent glass that had a transmittance of 88%.

METHODOLOGY
To study the effect of the opening distribution area on daylight performance at deep plan-private office space, the experimental method with simulation as a tool was used. A radiance-based computer simulation that had been validated in previous research (Canziani et al., 2004 andCourret et al., 1998) was employed. Radiance is a daylighting simulation program that uses a ray-tracing methodology to predict daylight's behavior in space accurately (Canziani et al., 2004). Characteristic of the materials used in this experiment was described in Table 1.
Daylight level and uniformity of the base case, HLP with an opening distribution area of 6.6 m 2 were compared with the case, HLP with an opening distribution area of 3.41 m 2 , simultaneously with daylighting standards. The base case and case had one opening distribution and two opening distributions, respectively. Those opening distributions were located at the center of each private office space ( Table 2).
The location of measurement points inside private office spaces can be observed in Figures 4 and 5. Twelve measurement points were located in each private office space. The measurement points had a distance of 0.50 m from the wall and had a distance of 1 m between one another. The height of measurement points was 0.75 m above the floor plan.
The experiment was carried out under overcast sky conditions in Surabaya (7º15' South Latitude and 112º44' East Longitude). Illuminance and Daylight Factor were simulated on 21 June at 09.00, when HLP's supplementary illuminance was significant (Chirarattananon et al., 2000).

RESULTS AND DISCUSSION
Daylight performance analysis was done by comparing illuminance level, Daylight Factor (DF), and uniformity ratio of the cases, simultaneously with daylighting standards. Horizontal illuminance guidelines by IESNA in Steffy, 2002 were used as daylighting standards. Horizontal illuminance target value for work area where simple visual tasks are conducted is 135-165 lux. Illuminance uniformity on workspace should be 3:1 on average to minimum and 6:1 in maximum to a minimum.

Illuminance Level and Daylight Factor
The results showed that HLP with an opening distribution of 6.6 m 2 (the base case) introduced an average work plane illuminance level as big as 127.1 lux. This illuminance level was below the illuminance target value for working space where a simple visual task is performed (135-165 lux) (Steffy, 2002).
Previous research by Mogo (2005) studied 9 m HLP under the same sky condition, overcast sky, but in higher latitudes (30°36' N). Slightly reduction of illuminance value compared with Mogo's light pipe occurred in this research. The reduction occurred because the HLP in this research was longer than Mogo's HLP.
A reduction of 50% of the opening distribution area, from 6.6 m2 to 3.41 m2 improved the daylight level. The case (HLP with an opening distribution area of 3.41 m2) introduced a higher average work plane illuminance level than base case (HLP with an opening distribution area of 6.6 m2). The average work plane illuminance level of the case reached 135.3 lux.
Compared to standards, the average work plane illuminance level performed by HLP with the opening distribution area of 3.41m2 had met the illuminance target value for working space where simple visual tasks are performed (135-165 lux). That office space can accommodate several activities such as casual reading, copy room, or as a computer-intensive office (Steffy, 2002).
The analysis also performed on illuminance value at all measurement points inside space to investigate the daylight performance of HLP with different opening distribution area thoroughly. Figure  6 shows that illuminance value on all measurement points with HLP's opening distribution area of 6.6 m2 was below the illuminance target value for the work area where a simple visual task is conducted (135-165 lux). A single HLP that had an opening distribution area of 6.6 m2 could not function as a working space where simple visual tasks are performed.
Different results occurred at office space with HLP's opening distribution area of 3.41m2. Illuminance value at most of the measurement points inside office space was in the range of the illuminance target value for the work area where a simple visual task is conducted (135-165 lux). That improvement of illuminance level occurred mostly on deeper office space, at a distance of 6-10.5 m from the side window.
(a) (b) = illuminance value < 135 lux Figure 6. Illuminance value on private office space with HLP's opening distribution area of (a) 6.66 m2 and (b) 3.41 m2 Figure 7 shows the percentage of measurement points that had illuminance value under and above the illuminance target value for the working area where simple visual tasks are conducted (minimum 135 lux). For the base case, the illuminance level on all measurement points inside space was under 135 lux. This result indicated that the office space could not be functioned as a working space where simple visual tasks are performed.
Different results appeared in the case, where the illuminance level on 62% measurement points inside rooms was above 135 lux and had met the illuminance target value for working space where simple visual tasks are performed (135-165 lux). Those measurement points mostly located in deep office space, at a distance of 6-10.5 m from the side window.
HLP with an opening distribution area of 6.6 m2 introduced average DF as big as 1.27% while HLP with an opening distribution of 3.41 m2 introduced average DF as big as 1.35%. These values were below the typical minimum DF for offices, as big as 2% (Lechner, 2015).

The Effect of Opening Distribution Area on Illuminance Level and Daylight Factor
The results indicated that with the same quantity and length (10.5 m), HLP with a smaller opening distribution area placed at the center of the space had a higher average illuminance level and DF than HLP with a larger opening distribution area (Figure 8). A 50% reduction of HLP's opening distribution area increased the average DF as much as 6.42%. Improvement of average illuminance level and Daylight Factor (DF) of HLP with a smaller opening distribution area is a new finding. These results showed a different tendency with previous research conducted by Beltran et al., 1997 about HLP's opening distribution area. Improvement of daylight levels in previous research was achieved not only by enlarging the opening distribution area but also by adding side reflectors and applying a trapezoidal section of HLP (Beltran et al., 1997). In this research, without any change in HLP's reflector and section, a 50% reduction of opening distribution area increased the daylight level as much as 6.42%. The presence of highly specular material on the opening distribution area had a contribution in specular reflection of daylight before being transmitted to office space by a translucent glass. Figure 9 indicates the DF profile (%) of the base case and case at the center of the office space. HLP with a smaller opening distribution area (3.41 m 2 ) had a higher DF level than HLP with an opening distribution area of 6.66 m 2 . The improvement of the Daylight Factor (DF) value of HLP with a smaller opening distribution area was in the range of 5.6% to 11.4%. The results showed the role of a highly specular material in reflecting the daylight before being transmitted to the office space. The profile also shows that in general, DF from Horizontal Light Pipe increased towards the interior of the room (Figure 9). This results had a good agreement with the previous research by Chirarattananon et al., 2000 about the pattern of daylight level from a light pipe. Figure 9 also showed that improvement of DF level occurred in the area adjacent to the back walls. This improvement reached because the light came from opening distribution was reflected off the wall and increased DF level near to it, in accordance with the results of Beltran et al., 1997 study.
Further observation of the DF profile under HLP's opening distribution area showed that the DF profile of HLP with an opening distribution of 6.66m 2 was similar, i.e. increased towards the back of each office space. Different DF patterns resulted from HLP with an opening distribution of 3.41 m 2 . DF profile of deeper office space, at the distance of 7.5-10.5 m from side lighting was higher than office space at the distance of 4.5-7.5 m from side lighting. The shape of HLP which tapers out towards the back of the space and the presence of highly specular material on the opening distribution area had a role in improving the DF level.

Daylight distribution
Analysis of daylight distribution was conducted by comparing the illuminance uniformity ratio between the base case and case, simultaneously with the recommended illuminance uniformity ratio on the workspace (Steffy, 2002). Illuminance ratio, consist of maximum-to-minimum and average-to-minimum were used to quantify lighting uniformity. Illuminance uniformity on workspace should be 3:1 on average to minimum and 6:1 in maximum to minimum (Steffy, 2002).
Both cases had a high uniformity ratio on space. Illuminance uniformity ratio, average to a minimum, was 1.07:1 for both cases. Illuminance uniformity ratio, maximum to minimum, were 1.12:1 and 1.15:1 for base case and case, respectively. Illuminance uniformity ratio of HLP with an opening distribution of 6.6 m2 and HLP with an opening distribution of 3.41 m2 was in the range of recommended illuminance uniformity ratio on the workspace.
These results indicated that as a main source of daylight, a single HLP running along the centerline of private office space could illuminate the space uniformly. HLP provided uniform daylight, not only as a complement to side lighting (Beltran et al., 1997 andCanziani et al., 2004) but also as the main source of daylight inside the space. HLP in private office space could function as a working space where a simple visual task is performed, such as computerintensive offices.

The Effect of Opening Distribution Area on Daylight Distribution
The base case had a higher illuminance uniformity ratio than the case. The results indicated that HLP with a larger opening distribution area distributed daylight more evenly than HLP with a smaller opening distribution area ( Figure 10). Both cases had a uniform daylight distribution, but the case had a higher daylight level than the base case. HLP with an opening distribution of 3.41m 2 then can be applied as the main source of daylight on office space. Figure 11 describes illuminance distribution patterns of HLP inside office space. Both cases had a similar illuminance pattern where all interior walls had a significant role in reflecting off daylight comes from opening distribution. Previous research by Beltran et al., 1997 showed that the back wall had a significant role in room illumination through HLP. The light that came from opening distribution was reflected off the wall and improved work plane illuminance near to it. However, in this research other walls also have the same role as the back wall (West wall) in room illumination, especially the front wall (East wall). This was because the room proportion studied in this research is smaller, that were 3 m in width, 4 m in length and 2.75 m in height. With a small proportion of the room, all walls had a role in reflecting daylight from opening distribution on the ceiling. The bright wall will make the room appear larger and more cheerful (Lechner, 2015). The closer the distance of the interior wall with opening distribution, the higher the illuminance contour resulted. In this study, the East and West walls which were perpendicular to the opening distribution had a strong role in reflecting daylight into space. The side walls (North and South walls) had a flatter illuminance contour, due to a greater distance from HLP's opening distribution (1-1.5 m).

CONCLUSION
Modifying opening distribution area will change the reflection and distribution of daylight inside Horizontal Light Pipe. A 50% reduction of HLP's opening distribution area, which means reducing the area of transparent glass and adding the area of reflective specular material inside the pipe is proposed in this research. The current research has investigated the impact of HLP's opening distribution area on daylight performance at deep plan private office space in the topics. A 50% reduction of HLP's opening distribution area, from 6.6 m2 to 3.41 m2 improved the average Daylight Factor (DF) level reached 6.42%. The presence of highly specular material on opening distribution areas contributed to the specular reflection of daylight before being transmitted to office space by a translucent glass.
In small room proportion (3 m in width, 4 m in length and 2.75 m in height), HLP along the centerline of room distributed daylight uniformly. A lower illuminance uniformity ratio but still meet the recommended illuminance uniformity ratio on workspace was resulted by a smaller HLP's opening distribution area. Considering the improvement of daylight level and high uniformity ratio resulted, HLP with a smaller opening distribution area can be applied as the main source of daylight on deep-plan office spaces.
Further research and development of HLP to meet the requirement for a more complex visual task in deep plan office space are needed. Several modifications of HLP's opening distribution element can be investigated, including the addition of reflectors and louvers.