0340~5442/93 $6.00+ 0.00
Energy Vol. 18, No. 2, pp. 197-210, 1993
ENERGY-EFFICIENT LIGHTING IN THAI COMMERCIAL BUILDINGS JOHN F. BUSCH+ Energy Analysis Program, MS 90-4000 Lawrence Berkeley Laboratory, Berkeley, CA 94720, U.S.A. PETER
International Institute for Energy Conservation, Racquet Club Building 8 Sukhumvit Soi 49/9, Bangkok 10110, Thailand SURAPONG
Asian Institute of Technology, GPO Box 2754, Bangkok 10501, Thailand
Abstract-We explore the opportunities to reduce the electricity required for lighting in prototypical Thai offices, hotels, and shopping centers. Using a whole-building energy analysis approach, we calculate the savings from lighting conservation measures directly, and from associated reductions in cooling load. Ancillary cost savings of air-conditioning energy and capacity comprise a significant 30 to 50% of the total. Lighting technologies considered in this study include electronic ballasts; @i-phosphor, narrow-diameter lamps; specular reflectors; occupancy sensors; lumen maintenance and daylighting controls; and compact fluorescent lamps. From a societal economic perspective, the cost of conserved energy of employing all applicable lighting conservation measures to these buildings is one-quarter or less than the average retail price of electricity for commercial customers, with a payback time of less than one year in hotels and retail buildings and approximately three years in offices. The internal rate of return for installing all lighting measures is 35% in offices, 142% for hotels, and 107% for shopping centers. Several current policy initiatives in Thailand, including a proposed energy standard for commercial buildings, with specific provisions for lighting, and plans by the nation’s electric utility to develop two commercial sector conservation programs, should help to spur the adoption of more efficient lighting products. We strongly recommend that the currently high tariffs and taxes (50% and greater) for imported efficient lighting products be reduced.
In recent years both general economic growth and commercial building construction in Thailand have boomed. Most new buildings are designed for a high level of amenity, including air-condi-
tioning and lighting systems that provide illumination levels comparable to those found in buildings in the West. This contributes significantly to the 15% annual growth in peak electricity demand for the country.S Designing and retrofitting buildings to use less energy is a way to avoid both high energy bills for building owners and the strain of rapid growth on the nation’s electricity infrastructure. In 1987, lighting accounted for 31% of electricity use in Thai commercial buildings.’ This enduse, therefore, is a good candidate for conservation investments. Many technologies exist that reduce energy consumption for lighting without sacrificing lighting services. These technologies generally add to initial costs for a lighting system, but their adoption can be justified by reduced electricity bills. In a hot and humid climate such as Thailand’s, these savings are derived from two areas. Direct savings come from reduced lighting consumption. Indirect savings come from reduced cooling loads due to the reduction in waste heat generated by the lighting system. t Author for correspondence. $ The demand for electricity from larger commercial buildings in the Bangkok metropolitan area is projected to increase more than threefold, to nearly 13,000 GWh, by tbe year 2001. During the same period, electricity use of large commercial buildings outside of Bangkok is expected to increase by a factor of 2.5, to 3,000 GWh.
F. BUSCH et al
In this article, we assess electricity conservation for lighting in an unusual context: a developing, tropical country that has rapid economic growth, relentless cooling demand, and economic policies that are geared towards promoting indigenous industries, but that can also impede the adoption of new, more energy-efficient technologies. METHODOLOGY
To analyze the savings potential in commercial building lighting systems, we follow a threestage approach. First, we evaluate lighting system design in detail, using a standard design methodology. Second, using a state-of-the-art building energy simulation program, we conduct a whole-building energy analysis. Third, we evaluate the economics of various lighting conservation measures based on the savings estimated in whole-building analysis and using local cost figures. Below we describe these three stages of the analysis. Fhorescent
Lighting System Design
Fluorescent lighting design calculations are based on the standard “lumen” method, as given in the Illuminating Engineering Society (IES) handbook2 and applied in a recent study of electricity conservation potential in Califomia.3 This methodology takes a specified design footcandle level, and, given assumptions about room geometry, room surface reflectivities, and various technical characteristics of lamps, ballasts, and fixtures, describes the number of fluorescent lighting fixtures required, the appropriate spacing, and the resultant installed power density in W/m2. Whole-Building Energy Analysis
The principal advantage of using the whole-building energy approach in assessing lighting efficiency options is that it can capture the interactive effects of lighting with heating, ventilating, and air-conditioning (HVAC) system performance. Since these interactive effects can be substantial, particularly in a cooling-dominated climate, this approach provides a more accurate basis than simplified end-use calculations for energy and economic analyses. Additionally, the whole-building energy approach provides a more precise accounting of the impacts on peak demand and the associated demand charge reductions that conservation measures can bring. This approach draws on an earlier study of conservation potential in the large commercial building sector of Thailand.4 Three elements form the core of the whole-building analysis: building prototypes, weather data, and a computer simulation model. Building Prototypes - We are principally concerned with air-conditioned
buildings because lighting systems, in giving off heat to the interior space, increase the energy required for cooling. The savings potential in lighting conservation is thus higher in these buildings. Also, the level of amenity implied by the presence of air-conditioning means that the lighting system has been professionally designed and, therefore, the conservation options we assess will have applicability. We focus on offices, hotels, and shopping centers, because more than 80% of large air-conditioning systems installed nationwide are found in these buildings. Prototype building descriptions are based on an actual building of that type in Bangkok and benchmarked to its historical energy consumption. To reflect typical current construction practice, we modified the building descriptions slightly. Characteristics of the prototypes are summarized in Table 1. The end-use breakdown for the buildings is as follows: lighting consumes 30% of the total energy for the office prototype; cooling and ventilation use 40%; and office equipment and elevators use 30%. For the hotel prototype, lights use 25% of electricity, while cooling and ventilation consume 60%, and elevators and miscellaneous equipment use the remaining 15%. In the retail prototype building, energy use for lighting is the largest share, at roughly 55%, followed by cooling and ventilation at 40%. The remaining 5% is shared by escalators and miscellaneous uses. Weather Data - Thailand has a hot and humid climate with year-round cooling loads and essentially no heating requirement. Temperatures vary within a limited range throughout the year, with the average dry-bulb temperature ranging from 25.5” C in December to 29.7” C in April. Hourly
Table 1. Characteristics
Thai commercial buildings
of Thai building prototypes.
Conditioned floor area Number of stories Number of guest rooms
6439 mt 12
20,628 m* 12 280
Single pane Tinted reflective bronze SC = 0.34
Single pane Tinted blue SC = 0.40
Single pane Tinted grey SC = 0.35
Perimeter: 14 mz/person Core: 6.5 mZ/person
2300 people (max.)
Mon-Fri: 8am-5pm; Sat: lam-noon
Public: 37 W/m2 Guest rooms: 10 W/m*
Occupied hours Lighting power density
Constant volume, distributed AHU
I Thermostat setting
Circulation: constant volume Shoos: solit-svstems
Public: constant volume Guest rooms: two-pipe fan coil
Shops: 74 W/m*
Supply fan capacity
13,152 lit/set (circulation)
Outside air quality
Water-cooled centrifugal chillers
Water-cooled centrifugal chillers
Air-cooled reciprocating chillers
data from Bangkok
for the simulation
as temperature, humidity, wind speed, and solar radiation.+ Simulation Model - We used the DOE-2.1D building energy simulation program to simulate the response of the building prototypes to changes in the lighting system under Thai weather conditions. The program solves the mathematical relations governing the thermodynamic behavior of a building on an hour-by-hour basis5 via four sequential modules: LOADS, SYSTEMS, PLANT, and ECONOMICS. In the LOADS module, based on user input describing the building surfaces, enclosed spaces, internal usage, and schedules, the instantaneous heating and cooling loads are calculated and then modified to incorporate dynamic effects of thermal mass through the use of weighting factors. The SYSTEMS module calculates the heat extraction/addition of the coils from a large menu of system types and operation parameters. Energy requirements to operate the primary heating and cooling equipment and pumps are determined in PLANT. The ECONOMICS module calculates the energy costs of operating the building and can accomodate complex tariff structures. Economic Analysis
Table 2 on the next page shows the component costs of lighting systems for the base case and efficient alternatives that are used in the economic analysis. For imported goods, we present the applicable import duties and business taxes, and their ultimate local market prices in Thailand. However, in keeping with a societal cost perspective in which taxes are viewed as a transfer payment and not a true social cost, we use only the pre-tax prices in our economic calculations. Furthermore, we employ only the incremental cost between the base t Weather data for 1985 were used, primarily because of the availability of good solar data for that year. Monthly average temperatures in 1985 had close correspondance to the 30-year norms.
JOHN F. BUSCH
Table 2. Costs of lighting components.’
Pre-tax price (baht)
duty & business taxes (%)
Local market price (baht)
1550 100 45 13
0% 0% 0% 0%
1550 100 45 13
55 952 57 750 250 875 226
55% 89% 50% 50% 55% 55% 55%
85 1800 85 1125 388 1356 350
Impon Lighting component Base Base Base Base
luminaire with acrylic diffuser magnetic ballast (l-tube, high power factor) fluorescent lamp (4-ft. F40) incandescent lamp (60W)
Tri-phosphor T8 lamp (32W) Specular parabolic luminaire Magnetic 32W ballast (l-tube) Electronic ballast (2-tube) Occupancy sensors (per fixture) Daylighting controls (per fixture) Compact fluorescent lamp (15 W, Integral)
case component and the efficient component. In that sense, the economics are oriented towards new or major retrofit situations, rather than minor retrofits, in which the full cost of the efficient component would have to be borne. For cost-effectiveness calculations, we assume a 6% real discount rate and a 20-year time horizon. Because some lighting system components have shorter lifetimes, we include the costs incurred for periodically replacing worn-out components up to the 20-year time horizon, and discount them from the time they are borne in our levelized investment costs. We assume a labor rate for changing bulbs of $0.20 per bulb. The costs for chillers and air-distribution systems per unit capacity were obtained from a consulting engineer in Bangkok who is familiar with local conditions.6 Energy cost savings are calculated based on the electricity tariff schedule for large commercial customers located in the Bangkok metropolitan area. The electricity tariffs have both an energy component at 1.23 baht/kWh (S$/kWh) and a demand component at 229 baht/kW/month (U.S. $9.16/kW/month). All monetary values (either U.S. dollars or Thai baht) given in this article are in 1991 units. With these assumptions we calculate three economic figures of merit: the cost of conserved energy (CCE), simple payback time (SPT), and internal rate of return (IRR). CCE is useful to those who prefer comparing a conservation investment to the typical price of electricity, or its expected future price. SPT is useful to those who have a rule-of-thumb about the threshold number of years required to pay off an attractive investment. IRR is appealing to those who evaluate a range of investment possibilities with different risks and rewards that can be reconciled in a percentage return on investment. LIGHTING
Different building types house different activities, as is reflected in the mix of lighting systems. Unfortunately, a comprehensive survey of lighting systems and their components has not been conducted for Thai commercial buildings. However, data collected in past building energy audits and interviews held with practicing lighting designers enabled us to develop a simplified profile of typical lighting systems installed in Thai offices, hotels, and shopping centers.‘J A typical fluorescent lighting system consists of a three-tube fixture with 40W bulbs (F40), high-power-factor core-coil ballasts, and a diffuser. Depending on the building type and vintage, fixtures may be suspended or recessed, and they may or may not have louvers and diffusers. As an intermediate configuration, we assume the base-case fluorescent fixture to be recessed and t Approximateexchange rate is 25 baht per 1 U.S. dollar.
Energy-efficient lighting in Thai commercial buildings
unvented, and have an acrylic diffuser, but be without louvers. We assume that the fluorescent fixtures are arranged to provide 50 footcandles of illumination and, when combined with the 150 watts-per-fixture and 3,000 lumen-output-per-lamp ratings, result in a base case power density of 22 W/m.2 We assume that all incandescent lighting in the base-case is provided with standard 60W bulbs. In fact, a variety of incandescent wattages are installed in the three building types. For the purpose of calculating the savings and cost-effectiveness of an alternative technology, however, no distortion is introduced by this simplification. All lighting in the base-case Thai commercial building prototypes are assumed to be comprised of either fluorescent and incandescent types in the following proportions (by total installed wattage): offices have 100% fluorescent; hotels are 30% fluorescent, 70% incandescent; and retail spaces are 45% fluorescent, 55% incandescent. Lighting Conservation Measures Good reviews of the technical options for increasing the efficiency of lighting systems can be found elsewhere.g-” We considered the following measures for increasing the efficiency of fluorescent lighting systems: (i) electronic ballasts; (ii) u-i-phosphor, narrow-diameter (T8) lamps; (iii) specular reflectors; (iv) occupancy sensors (for offices only); and (v) lumen maintenance and daylighting controls (for offices only). For incandescent lighting we analyzed standard bulb replacement with compact fluorescent lamps (CFLs) as the single savings option. RESULTS
A sample printout of the computer spreadsheet model is shown for the prototypical hotel as Table 3 on the following two pages. The rows, beginning at the top, first list the parameters, such as the footcandle design level and discount rate, then design calculations for the fluorescent lighting, then incandescent lighting, and, finally, the whole-building analysis, which includes the savings and the economic performance calculations. Where calculations are performed, they are given next to each line item. The columns in Table 3 begin with the base case lighting system; conservation measures are successively added to that system with each column as one moves to the right across the table. In other words, the analysis is not parametric, but cumulative. Thus, the last column in Table 3 under the heading “Compact Fluorescents” connotes a system that incorporates all of the conservation measures together. Table 4 shows the electricity savings that result from applying all the lighting measures. The first set of columns present savings as a percentage of the base case consumption of each enduse. By implementing all lighting measures applicable to each building type, nearly 70% savings in energy for lighting are possible. In the cooling and ventilation end-uses, the savings are more modest but still significant. In retail buildings, lighting forms such a large portion of total electricity use (much higher than in hotels or offices), that when large savings in lighting are produced, large savings in air-conditioning necessarily follow. For this reason, cooling and ventilation savings in the retail building are as high as 32 and 25%, respectively. This also explains why retail buildings save 48% of total electricity consumption from the full set of lighting efficiency measures, while offices and hotels save half as much proportionally (25 and 27%, respectively). While not shown in the table, peak savings are comparable on a percentage basis to the energy savings. The second set of columns in Table 4 reveal the allocation of total savings among lighting, cooling, and ventilation.+ There is quite a difference among building types in the percentage of total savings that come directly from lighting, ranging from 64% in the case of the hotel to 81% in the office.
t “Cooling” describes the energy use of the chiller alone. “Ventilation”refers to the air-distributionsystem of fans, ducts, pumps, and piping that together deliver cooling to zones that demand it.
_. .” ._ . .
4. Levelircd lamping + relamp 1. Total levelizcd cost/fixture (FC4+VC3]
I. Marginal lamp cost 2. Total lamp cost [VCZpre+VCll 3. Adjusted 10 base case [VCZ*Lt6base/Lt61
Lamp cost (VC)
1. Marginal Fixture Cost 2. Total fixture cost [FC?pre+FCll 3. Adjusted to base case [FC2*Lt6base/Lt61 4. Levelized [pmt(P4.P5.-FC3)]
4. Lamp life (hm) 5. Relamp interval [0.7’L4/(C1*C2*pJ*8760)1 I. Lamps per fixture 2. Fixture Wsltage 3. Coefficient of utihzalion 4. Ballast factor 5. Din depreciation factor 6. Tmp. and volbge nuctuatlon faCtO1 1. Light loss factor [F4*FS*F61 2. Lumens per fixolrc [F3*LZ*L3’Fl*Ltll 3. Number fixtures nquired [PI*P2/LtZl 4. Average fixture spacing [(PULt3)‘%51 5. Power density [W’FUPII 6. Finure coverage lPl/LDI I. Adjust for Occupancy Sensors or S&d. 2. Adjust for lumen maint. and daylighting
1. Wattage 2. Rated lumens 3. Lamp hnnm depreciation factor
Fixture Cosl (FC)
40 3ooo 0.84 zoo00 3.2 3 150 0.7 0.94 0.81 0.95 0.72 3828 946 8.7 2.0 77 1.0 1.0 SO.00 $74.00 $74.00 $6.45 SO.00 $1.80 $1.80 $2.07 58.52
32 zwo 0.86 zoo00 3.2 3 126 0.73 0.94 0.81 0.97 0.74 4034 897 9.0 1.6 81 1.0 I.0 SO.00 S74.00 $70.22 56.12 $0.40 $2.20 $2.09 $2.37 $8.49
Tri-phosphor T-8 lamps
32 2900 0.86 20000 3.2 2 84 0.9 0.94 0.81 0.97 0.74 3316 1092 8.1 1.3 66 1.0 1.0 $4.08 $78.08 $90.15 57.86 $0.00 $2.20 52.54 $1.89 $9.75
Table 3a. Design calculations, savings, and economics of electricity conservation for lighting in Thai hotels.
32 2900 0.86 zoo00 3.2 2 62 0.9 0.94 0.81 0.97 0.74 3316 1092 8.1 0.9 66 1.0 1.0 $22.00 5100.08 %I 15.54 510.07 $0.00 $2.20 $2.54 $1.89 $11.97
1.IRR [rateQ5,S9.(AC2bue-AC2))1 2. lRR w/HVAC uvinns[[email protected]
18% 4406 154622 2290 $516.032 $540.122 8.5
60 860 0.93 1000 0.2 0.7 0.8 448 SO.52 $3.28
1759 4321 151433 2264 S500,%3 5524.613 7.9 137 85 222 $1,352 $4.140 $5.492 $479 $15.069 (SO.ca2) (so.~) ($0.004) 0.0 0.0
Tli-phosphor T-8 lamps
1656 4258 149042 2244 5489665 S516.342 7.5 240 148 388 52.366 $7.324 $9691 $845 526.367 $0.018 SO.007 $0.004 1.1 0.7 91% 13740
1519 4175 145853 2218 5474,606 5503.700 6.8 377 231 608 53.718 $11.464 $15.183 $1.324 $41.426 SO.01 1 $0.008 $0.006 1.4 1.0 73% lcn%
Note: Calahtions TICcumulative in movingfrom left to rightalongthecolumns(e.g..“CompactFlourescent” includesresultsof tri-phosphur T-8 lamps,specuhrreflectOn.andekClmiC
6. Chillercostsaving&2 [(Ehwe-E4)*P8*3413/12CKtO) 7. HVAC costsavings[S6+S7] 8. LevelizcdHYAC costsavings[[email protected]
,p5.S81 9. B”Wgy C‘[email protected]
] 1.hlarginllCCB[(TC-Tqxe)/(scs4pre)l 2 Avenge CCB [(-l-C-K!base)/S41 3. Avenge CCE wl HVAC swings[(TC-TCbase-SS)/S4] 1.Simplepaybck [(AC2-AC2lmse)/S91 2 SPT WIHVAC savings[AC?-AC2base-S7lS9]
1.Lkwnll ==ly (Mwh) 2 HVAC energy(MWh) 3. Fanupcity (litbec) 4. Cltillcropacity (Irw) 5. bergy cost 6. LevelizedIiglmittg ittvertment andoperatingcost[ACl+ESl 1.Lightingatugy “se wP91 wwft2) 2 Lightingatetgy swings [Elbnre-El] 3. HVAC atetgy savingsFpUe-E21 4. Totalenergyuviugs [S?+S31 5. Fancostwings [(E3bw-B3)‘I’7*6Of28.3)
ax-2 Emcm (E)
1.Levdized totallightingcat [(TCI*Lt3) + (KPP8)l 2. Totallightingfirstcost&FC3+ VC3*Fl)‘Lt3 + ICIW3)
1.hnp cost 2. Leveliaxl lmping & rthtping cost/lamp
1.wattage 2RatedLummr 3.Lmnpluna depreciation factor 4. Lamplife (ltoun) 5. Relanpinginterval[IL4/@‘3*8760)1 1.Codficiea ofutilitutia, 2. Dirtdepcciuiattfauor 1.Adjustedlumens[IL2*IL3*LLl*LLZ]
Table 3b. Design calculations, savings, and economics of electricity conservation for lighting in Thai hotels.
6u 368: 12792: 2w $376.004 s411,151 2: 128! 72’ 2011 511,320 539.488 $50.808 $4,430 Sl40.028 $0.004 SO.006 $0.003 0.’ 0.: 1424 2954
1: 9a 0.8: looo( 2.: 0.8: 0.1 521 $9.04 $4.51
F. BUSCH et al
Table 4. Electricity savings from fnll lighting measures. -l
I End use
As a percentage total
As a percentage of end-use Office
Lighting Cooling Ventilation
Cooling savings account for 13 to 23% of the total savings; ventilation accounts for 2 to 15% of the savings. The reduction in lighting energy for hotels leads to substantial reductions in the cooling load. This is because hotels, or rather their air-conditioners, operate 24-hours per day every day of the year. Therefore, all cooling loads developed within a hotel must be met by the air-conditioning system. By contrast, in offices or shopping centers, some heat generated during operating hours dissipates at night and never places a cooling demand on the air-conditioning system. This is a compelling demonstration of the importance of considering lighting conservation potential by building type, as opposed to a more aggregate approach. Studies of lighting savings conducted in more temperate climates have shown a much smaller bonus from savings in cooling and ventilation. l2 This is partially due to the relative severity of the cooling loads in the tropics. But it is also due to the fact that air-conditioning systems in Thailand tend to be less energy efficient. Constant-volume air distribution systems are the norm there, and continue to be specified over the more efficient variable-volume technology. Only rarely are the more efficient chillers, common to the U.S., installed. In short, in circumstances like those of Thailand, the failure to augment the direct savings from lighting conservation with the ancillary savings of air-conditioning will lead to gross underestimates of conservation potential and cost-effectiveness. Figure 1 shows the economics of lighting conservation in Thai commercial buildings. The figure is divided into sub-charts depicting the three cost-effectiveness indicators CCE, SPT, and IRR. Though the economic results are averages based on measures added cumulatively, each individual measure’s impact on cost-effectiveness is apparent by observing marginal changes in the figure. The lower group of bars in each of the sub-charts give explicit credit for savings in the capital cost of HVAC equipment which have been downsized due to the smaller cooling loads resulting from lighting conservati0n.t As a practical matter, these savings are idealized; WAC equipment comes in discreet sizes and it is unlikely that the full amount of the savings reflected in the HVAC downsizing credit could be captured in an actual building. Still, the juxtaposition indicates potential savings. Overall, the HVAC downsizing credit causes about a 10% improvement in cost-effectiveness. The most meaningful comparison for the CCE in Fig. 1 is with the average retail cost of electricity. Because the electricity tariff for the large business customer class in Thailand is structured with both energy and monthly maximum demand charges, and each of the building prototypes has a unique pattern of energy and peak demands, the CCE is best compared to the average unit electricity cost specific to that building prototype.+ For offices, the average electricity price is $0.087 while the CCE for full conservation measures is $0.019. Similarly, for hotels, the electricity price is $0.069, while the CCE is $0.006. And for retail buildings, the electricity price is $0.082, while the CCE is $0.007. Thus, the CCE of employing all applicable lighting conservation measures to these buildings is one quarter the cost of electricity in the worst case, and about one-tenth the cost in the other cases. The SPT for full lighting conservation measures ranges from less than one year in hotels and retail buildings to about three years in offices. Finally, the IRR of installing all the lighting measures is 35% for offices, 142% for hotels, and 107% for shopping centers. t The reahzation of a HVAC downsizing credit would be likely only in new applications or major retrofits. However, this is consistent with onr treatment in the economic analysis, $ Average unit electricity cost is calculated as the total annual electricity cost (including both energy and demand charges) divided by the electricity consumed annually.
Energy-efficient lighting in Thai commercial buildings
Office wl HVAC Sav. Hotel WI HVAC Sav. Retail w/ HVAC Sav. -1
CostofConwwed Ena~y (t/l&W)
Office WI HVAC Saw
Hotel WI HVAC Sav.
Retail w/ HVAC Sav. m 3. 0
Office WI HVAC Sav. Hotel w/ HVAC Sav. Retail w/ HVAC Sav. I
50 100 internalRate OfReturn(%)
nTri-phos. T-8 Lamp q]Spec ReRector
fl Elec. Ballast
Fig. 1. Economics of lighting measures (social cost perspective).
CFLs, electronic ballasts, and t&phosphor narrow-diameter lamps (T8s) prove to be the most economically promising technologies. Specular reflectors, and controls for occupancy and lumen maintenance/daylighting (the latter two assessed for offices only) are somewhat less attractive economically. Still, they are much cheaper on a marginal basis than avoided electricity costs.
JOHN F. BUSCH et al
As mentioned earlier, these economic results are indicative of a societal cost perspective. lndividual building owners currently see higher prices due to duties and other taxes on imported efficient lighting products. These increased costs ultimately depress the economic performance, as compared to the results shown above. Likewise, the cost of money (as expressed in a discount rate) for a building owner is most likely higher than the societal value of 6% applied in the calculations above. Nevertheless, even when using local market prices that include import duties and business taxes (shown in Table 2) and a discount rate more than twice as high as above, the essential cost-effectiveness of investment in lighting efficiency is not changed. POLICY
Despite the strong economic incentives to install energy-efficient lighting systems in Thailand, such systems remain the exception rather than the rule. Part of the problem has been a lack of information about the options, their savings, and the life-cycle costs associated with their use. The architectural and engineering community is just beginning to become aware of the opportunities for conservation in commercial buildings. Still, in Thailand’s very competitive market for commercial space, building owners and developers are often reluctant to consider any measures designers might propose that will increase initial costs. Moreover, the market for energy-efficient lighting products suffers from a “chicken-and-egg” problem: because the market is small, products are expensive and in short supply, which, in turn, ensures a small market. Fortunately, several initiatives to overcome these barriers are proceeding. Commercial Building Energy Standard
Thailand’s economic boom of recent years has led to huge increases in the stock of commercial buildings. In an attempt to control the growth in electrical demand from these buildings, the Ministry of Science, Technology, and Energy has drafted a Building Control Act. The Act, which has not yet been circulated for public comment, will have separate guidelines for large (i.e. greater than 500 kW) and small (i.e. 30-499 kW) buildings. Broadly speaking, the code will establish standards for materials used in the building envelope, for heat transmission of the building envelope using the overall thermal transfer value formulation, and for lighting and air-conditioning energy use, The code also will set guidelines for energy intensity and building operation and control. The code probably will be introduced initially on a voluntary basis. In lighting, the code will assign illuminance levels for various tasks and efficacy ranges for different types of lamps and will establish a lighting budget with maximum lighting power intensities in W/m2 for various building types, as presented in Table 5. The lighting budgets in the standard for offices, retail buildings, and hotels range from 16 to 23 W/m2, two-thirds to one-quarter less than the installed lighting power density of typical existing buildings. For reference, we show in the last column of Table 5 the installed power density for lighting that results from applying the full complement of lighting efficiency measures to the building prototypes. The efficient prototypes improve upon the proposed standard by roughly one-third, depending on the building type and function within it.
Utility Conservation Programs
In 1990, the government directed the Electricity Generating Authority of Thailand (EGAT), the public enterprise mandated to generate and transmit electricity throughout the nation, to develop a plan for committing utility resources in pursuit of electricity conservation opportunities. The recently completed plan projects energy savings of 1080 GWh annually and peak savings of 225 MW, at a cost of U.S.$180 million, spanning a five-year period.13 The estimated cost of saved energy is 2.0 $/kWh, as compared to EGAT’s avoided cost of new supply of 4.3 c/kWh. Pilot programs to demonstrate energy savings are expected to begin in 1993. The government has agreed in principle that EGAT and the local distribution companies may recover their expenses on conservation programs through a tariff adjustment mechanism.
Energy-efficient lighting in Thai commercial
Table 5. Lighting budget in the proposed Thai energy standard.
Building type/space offices Hotels/motels
Retail Main concourse at multi-store shopping centers Food service
Maximum allowable lighting power (W/m?
Efftcient prototype, full measures (W/m?
Fine merchandising General merchandising
Fast food/cafeteria Leisure dining/bar
15 14 15 2
10 _ 5 -
16 18 18 5 18
Area types Guest rooms & corridors Public areas Banauet and exhibit
Garages and basements Schools
Pm/elementary High school TechnicalAmiversities
Warehouses/storage Hospitals/nursing homes
Two types of programs have been targeted for the commercial sector: (i) improving the overall efficiency of new commercial buildings; and (ii) increasing lighting efficiency in existing and new commercial buildings. The development of utility lighting conservation programs over the next several years in Thailand can be expected to significantly reduce electricity use for lighting in Thailand’s rapidly growing commercial sector. Programs also will stimulate the market for efficient lighting products and services. The Market for EfSicient Lighting Products The market for fluorescent tubes in Thailand is about 35 million tubes a year, or roughly 900 million baht (U.S.$35 million). Three large manufacturers (Philips, Toshiba, and AsiaLamps Industry) account for about 82% of sales. The relatively small size of the Thai market makes it expensive for companies to retool and invest in new manufacturing capacity to produce more efficient products. The first company to introduce energy-saving fluorescent lamps began importing 36W tubes on a limited basis about five years ago. Recently, Philips entered the market with an aggressive promotion for its 36W and 18W tubes, which it began manufacturing in Thailand in 1990. By 1992, it had converted all of its production to the higher efficiency 18W and 36W tubes. At present, however, 36W, 37W, and 18W tubes are the only energy-saving fluorescent tubes available in Thailand. For reasons described in more detail below, energy-saving 28W, 32W, and 34W tubes are not available in the local market. Clearly if Thailand is to improve the efficiency of its commercial lighting in the short term, it will need to import efficient products tubes, ballasts, and compact fluorescents, at least until these products can be manufactured domestically. Yet there remain significant barriers to the importation of efficient hardware. The middle column of Table 2 shows the duty and business taxes imposed on lighting products imported from abroad (expressed as a combined percentage increase in the pre-tax price). Two highly efficient products that are not manufactured in Thailand are electronic ballasts and compact fluorescent lamps.? The high import duties and taxes effectively inhibit the import of electronic ballasts and slow the adoption of CFLs. Nevertheless, the market for imported CFLs has begun to grow dramatically in recent years. Most of the sales are of the modular type, which
manufacturers am planning to begin domestic production of CFLs.
have separate ballasts and tubes; these cost 140 to 190 baht (U.S.$6.00-8.00). Screw-in fluorescents cost 350 to 600 baht (U.S.$ 14.00-24.00). By comparison, a 75W incandescent bulb costs about 15 baht (U.S.$ 0.75). Still, the demand for CPLs is steadily increasing. Unlike the U.S., where the residential and commercial retrofit sectors form a large share of the CFL market, Thai sales are almost entirely directed to new commercial buildings, particularly hotels. Sales currently are about 500,000 units per year and growing by 30 to 40% annually. They are projected by some manufacturers to increase to one million units by 1993. Because of the 50% import duty and taxes, virtually all ballasts sold in Thailand are manufactured by about twenty manufacturers. Most produce low-power-factor ballasts with a power factor of 0.5 or less. These retail for about 35-40 baht (U.S.$l.40-1.60). About five Thai companies manufacture high-power-factor ballasts, which have a power factor of approximately 0.9, and retail for about 100-150 baht (U.S.$4.00-6.00). The cost of installing a capacitor in the fixture circuit is lower (70-80 baht, or U.S.$2.80-3.20) and is currently the design approach preferred by most electrical engineers in Bangkok. The standard 4oW magnetic ballasts sold in Thailand have losses of about 1OW. Yet exorbitant duties and taxes make importation of magnetic ballasts with lower losses or high-frequency electronic ballasts unattractive. The testing procedures required by the Thailand Industrial Standards Institute (TISI) for imported ballasts exacerbate this problem. Currently all magnetic ballasts have to be tested by TISI. In addition to meeting the standards of the International Standards Organization, TISI regulations specify that all ballasts go beyond safety standards to cover performance (e.g. current and power factor) relative to a reference ballast. If TISI has not established a reference ballast to one whose import is sought, technically it cannot be imported. As a case in point, an Australian company, Atco Co., produces a low-loss (4.5W) 32W magnetic ballast that can be used to power high-efficiency 32W lamps. This combination has significant advantages over the standard 40W ballast-lamp combination. Currently, though, the only 32W reference ballast in Thailand is for the “circline” type of lamp, which has different operating characteristics than the Atco 32W ballast, which is designed for straight tubes. For lack of a reference ballast, it is currently not legal to import the 32W Atco version. As a result of such policies and testing regulations, the most efficient locally available lighting configuration is a 36W tube with a 40W magnetic ballast. Potential Policy Reforms
There are at least two broad areas in which policies can be reformed to encourage use of higher efficiency lighting products in Thailand. These are (i) import duties; and (ii) test procedures and regulations. Import duties must be high enough to protect local industry while at the same time encouraging local manufacturers to work harder to improve the quality and efficiency of their products. Although the Thai government recognizes the need to encourage energy and environmental conservation, and has promoted several measures to achieve these ends, it has not addressed the issue of high import duties on lighting equipment, which make many efficient imported lighting products too expensive for the local market. In 1985, the Thai government instituted a system to allow for rebating a portion of import duties on new equipment that would save energy or reduce environmental pollution. The system allows for a reduction of up to 50% in the import duty for equipment intended for projects with a minimum value of U.S.$ 15,000; in no case, however, can the duty be reduced below 10%. To date, no applications have been submitted for this import duty reduction. Currently, the Thai government is considering a tax reduction that would lower import duties on components and materials used to produce energy-efficient products. In the lighting sphere, it appears that this reduction will not apply to products that can be manufactured in Thailand, but instead would apply only to those products (such as electronic ballasts and compact fluorescent lamps) that are not now manufactured in Thailand. Philips, which has set up a factory to assemble CFLs using imported components, recently applied for duty relief under the proposed tax reduction scheme but thus far has been denied. Finally, the Thai government is under pressure to comply with the guidelines set by the General Agreement on Tariffs and Trade (GATT). As a broad policy objective, GATT gets a target for
Energy-efficient lighting in Thai commercial buildings
the maximum import duty at 20%. For this reason, it appears likely that import duties on a variety of protected products (not just lighting) will be lowered over the next several years. The Thai government should take clear action to reduce the import duty on products that can save energy, as well as components and equipment used for their manufacture. TISI’s test procedures also need to be streamlined so that they maintain a commitment to product safety, but do not inhibit the adoption of new technologies, such as 32W magnetic ballasts or ballasts for compact fluorescent lamps. CONCLUSIONS
Efficient lighting technologies employed in commercial buildings in Thailand offer a large elec tricity conservation potential. A combination of efficient fluorescent lighting system components and compact fluorescent lamps in place of typical lighting in Thai offices, hotels, and retail buildings would save from one-quarter to one half the total building electricity use. Much of the savings derive from reduced energy consumption of the air-conditioning system when lighting-related cooling loads are lowered; these can account for as much as 35% of the total energy savings. The economics of investing in more efficient lighting also are extremely favorable. From a societal perspective, the cost of conserved energy for the full complement of lighting conservation measures considered here is one quarter or less than the average retail price of electricity charged to commercial customers. These obvious advantages have not translated into significant penetration of efficient lighting systems in the Thai market. One of the problems is the relatively high local market prices of efficient lighting products due to little local manufacture and the high duties and taxes imposed on those imported from abroad. It would behoove the Thai government to take steps to reduce this formidable barrier to rational energy use, first by reducing import duties and taxes, and then by establishing incentives for the manufacture of efficient lighting products locally. Meanwhile, the government has put forth two policy initiatives that should help spur the adoption of more efficient lighting products. These include a proposed energy standard for commercial buildings with specific provisions for lighting, and fresh plans by the nation’s electric utility to undertake conservation programs. Two of these programs target the commercial sector, with lighting as a key element in them. While these efforts are heartening, special diligence is required to capture even a small fraction of the available savings identified in this study. We believe it is worth the effort. REFERENCES
1. National Energy Administration, Ministry of Science, Technology, and Energy, Royal Thai Government, “Electricity Savings Plan for the Industrial, Commercial, and Residential Sectors,” Bangkok, Thailand (January 1990). 2. Illuminating Engineering Society of North America, IES Lighring Handbook, 1984 Reference Volume, New York, U.S.A. (1984). 3. D. Goldstein, R. Mowris, B. Davis, and K. Dolan, “Initiating Least-Cost Energy Planning in California: Preliminary Methodology and Analysis,” Submitted to the State of California Energy Resources Conservation and Development Commission, Docket No. 88-ER-8, Sacramento, CA, U.S.A. (February 1990). (Revised in Personal communication with Robert Mowris, May 1991). 4. J. F. Busch, Jr., “From Comfort to Kilowatts: An Integrated Assessment of Electricity Conservation in Thailand’s Commercial Sector,” Lawrence Berkeley Laboratory Report LBL-29478, Berkeley, CA, U.S.A. (August 1990). 5. B. Birdsall, W. F. Buhl, K. L. Ellington, A. E. Erdem, and F. C. Winkelmann, “Overview of the DOE2 Building Energy Analysis Program Version 2. lD,” Lawrence Berkeley Laboratory Report LBL- 19735, Rev. 1, Berkeley, CA, U.S.A. (August 1990). Boonpong Kijwatanachai, Engineer, MITR Technical Consultants, Co. Ltd. Bangkok, Thailand, personal communication (November 1989). Wiboon Luangviriyasang, Chief Electrical Engineer, Design Dept., Environmental Engineering Consultants, Bangkok, Thailand, personal communication (March 1991). Prasit Pittayapat, Asst. Professor, Dept. of Electrical Engineering, Faculty of Engineering, Chulalongkom University, Bangkok, Thailand, Personal communication (March 1991). S. Nadel, H. Geller, F. Davis, and D. Goldstein, “Lamp Efficiency Standards for Massachusetts: Analysis and Recommendations,” Massachusetts Executive Office of Energy Resources, Boston, MA, U.S.A. (June 1989).
JOHN F. BUSCHet
10. Mills and Piette, Energy -- The International Journal 18, 75 (1993). 11. California Energy Commission, “Advanced Lighting Guidelines,” Publication P400-90-014, Sacramento, CA, U.S.A. (March 1990). 12. A. Usibelli, S. Greenberg, M. Meal, A. Mitchell, R. Johnson, G. Sweitzer, F. Rubenstein, and D. Arasteh, “Commercial-Sector Conservation Technologies,” Lawrence Berkeley Laboratory Report, LBL-18543, Berkeley, CA, U.S.A. (February 1985). 13. International Institute for Energy Conservation, “Demand-Side Management for Thailand’s Electric Power System: Five-Year Master Plan,” Submitted to the Electricity Generating Authority of Thailand, Metropolitan Electricity Authority, and Provincial Electricity Authority, Bangkok, Thailand (November 1991).