Printer Friendly

Utility and consumer-oriented multi-criteria assessment of residential light bulbs available on the Australian market.

1. Introduction

The number of residential premises has been growing significantly over the last few years in Australia. As an example, the Housing Industry of Australia (HIA) has reported that 90-120 thousand dwellings have been built every year in Australia since 2010 ('Window into Housing' 2015). In some capital cities such as Sydney and Perth, over 30 thousand houses have been constructed in one year (Hennessey 2015; Duke 2016). Also, the average size of a new Australian house has increased by over 40% from 162.2 to 227.6 square meters between 1984 and 2003, which becomes almost 10% bigger than that in the United States of America (US). Different types of electrical appliances are used in residential premises; among which lighting load is the fundamental and unavoidable one. Considering the above factors, the lighting demand illustrates a growing trend in the Australian residential electricity sector (Sorensen 2013; Johanson 2011). Before 2011, Australian houses could consume up to 25 W electricity per square meter (W/[m.sup.2]) of their house floor space ('BCA lighting restrictions' 2017). Conventionally, incandescent light bulbs were the main source of lighting but not very efficient as they have only 16 lm/W on average and 95% of the consumed energy was wasted as heat (Wells 2013).

However, with the development of a new regulation on using energy efficient lighting systems in residential premises by the Building Code of Australia ('BCA lighting restrictions' 2017), this has been reduced to 5W/[m.sup.2] for indoors, 4W/[m.sup.2] for outdoors, and 3W/[m.sup.2] for garages. This new Building Code has significantly pushed the new house builders and those renovating (over 50% of a house) for using energy efficient lamps which have a minimum of 27 lumens per Watt (lm/W) such as compact fluorescent lamps (CFLs) and light-emitting diodes (LEDs) that have an average of 60 and 150 lm/W, respectively (see Figure 1) (Wells 2013; 'Energy efficient lighting...' 2013; 'Lighting Catalog. . .' 2013). The estimation of ('Phase out of. . .' 2017) shows that this will reduce Australia's greenhouse emissions by 800,000 tons, equal to a saving of approximately 0.14%. In a similar way, many countries in the world have already phased out the use of incandescent light bulbs (see Figure 2 (Kooroshy et al. 2015)) by adopting regulations on banning their production, import, and sale for general lighting purposes (Matvoz and Maksic 2008). As an example, incandescent light bulbs of 40 W and above are banned across the US since 2014 by which the nation's electricity consumption has reduced almost $10 billion every year (equal to the saving from 30 power plants across the country) (Bravo and Abed 2013). Following these bans, CFLs and LEDs have gained a large acceptance and interest among the people; even though their costs are slightly higher.

Based on 'How much electricity...' (2016; 'Household energy efficiency...' 2017), 7 and 10% of the total residential electricity demand in Australia and US, respectively comprises the lighting demand. The relevant energy consumption in Australian residential lighting sector is projected to approximately rise to 25 x [10.sup.15] J of energy by 2020 ('Energy use in...' 2008). In a similar trend, it is anticipated in 'Energy savings forecast...' (2016) that by 2035, LEDs will hold 86% of lighting installations in the US, compared to 6% in 2015. This will lead to an annual savings of 1.14 x [10.sup.15] J of energy in 2035 (almost equal to the total annual energy consumed by 45 million homes in the US today). It is also projected that the total savings by replacing classic incandescent by LEDs between 2015 and 2035 are equal to $140 billion. According to Yong et al. (2010), the International Energy Agency has estimated that changing traditional incandescent/halogen lamps to CFLs and LEDs would cut the world's electricity demand by 18%. Moreover, an increase in the number of LEDs can be very helpful not only to save power but also to solve global problems such as reduction of greenhouse gas emissions in developed countries by 25-40% in 2020 and by 50-85% in 2050 (Bessho and Shimizu 2012).

Although LEDs and CFLs bring significant savings in energy efficiency and equivalent costs, they are generally considered as non-linear loads versus standard incandescent lamps, which are linear. A compact ac-dc converter supplies a dc current to LEDs and CFLs, which introduces non-linearity to the system. This implies that the current drawn by these lamps do not have a sinusoidal waveform, even if they are supplied with a sinusoidal voltage (Cuk et al. 2010; Gil-de-Castro et al. 2014). As an example, the study in Bhattacharyya and Cobben (2011) shows that LEDs produce more current total harmonic distortion (THD) than refrigerators, TVs, and computers do which have a typical THD of 10, 48, and 87%, respectively. As reducing THD is very difficult once the system is polluted, it is more economical to install those LEDs and CFLs that have lower levels of harmonic injection (Watson, Scott, and Hirsch 2009). As an example, Lam and Jain (2008) reduces the current THD by modifying their valley-filled circuits, which also increases their power factor significantly (to approximately 0.98). To meet the challenges of manufacturing lamps with low cost and low harmonic emissions, some solid state-based ballast circuits may also be used (Meyer et al. 2017).

Even though each LED or CFL is a small load of a few watts, their accumulated impact may not be negligible on the distribution networks supplying the residential premises (Molina, Mesas, and Sainz 2014). Thus, problems can stem from the flow of non-active energy caused by harmonic currents and low power factor (Liang 2017; IEEE Std-1459, 2010). This may lead to voltage distortions, increased power losses, overloaded neutrals, transformers' heating and aging, unnecessary operation of protective relays, mal-operation of circuit breakers, and deterioration/failure of power factor correction capacitors (Fuchs and Masoum 2008; Arefi et al. 2012). Domestic customers may also complain about burning and failing of household appliances. With the advancement in technology, and increasing awareness of consumers for using energy-efficient lamps, on top of ongoing new governmental incentives, regulations and standards, these type of lamps are taking a larger share of the market (see Figure 3)('LEDs to account. . .' 2013). Thus, their adverse impact on the distribution networks needs to be evaluated.

This research has mainly focused on analysing the majority of residential light sources (i.e. CFLs, LEDs, and Halogen lamps) available on the Australian market. The study aims to assess these lamps against utility and consumer-oriented criteria. From the utility perspective, the paper provides a detailed comparison between different types of lamps, manufactured by different companies and available in different ratings, from power quality aspects including injected current harmonics and THD, as well as the active power consumption and power factor. A comparison of fundamental apparent power and non-fundamental apparent power for the lamps is also included. From the consumer-oriented aspects, the lamps are compared against their nominal luminous efficacy, cost and lifespan, provided by the manufacturers. Finally, a multi-criteria assessment of all studied lamps is presented to identify the overall performance of lamps from both utility and consumer-oriented perspectives. In summary, the main contribution of the paper to the research field is evaluating the residential lamps sold currently on the Australian Market considering both consumer and utility perspectives using a multi-criteria assessment.

The rest of the paper is organised as follows: Section 2 introduces different types of residential lighting and presents a brief overview of their operation mechanisms and characteristics. The utility and consumer-oriented criteria based on which the lamps are evaluated in this research are discussed in Section 3 while Section 4 introduces the methodology of the analysis and experiments. Section 5 presents the juxtaposition of the considered lamps from consumer-oriented perspectives while the results of their comparison from utility-oriented aspects are presented in Section 6. Section 7 presents the overall multi-criteria assessment of the studied lamps while the main findings of the research are highlighted in the last Section.
Figure 3. Residential sector light bulb purchases.

       Incandescent   CFLs   LEDs

2012   3,446          4,390     109
2013   6,317          7,011   2,819
2014   9,274         11,476   8,771
2015  11,285         15,497  12,004
2016  13,526         19,785  15,716

Note: Table made from bar graph.

2. Different types of residential lighting

The residential sector lighting lamps, sold on the Australian market, can be classified under the main three categories of LEDs, CFLs, and Halogen lamps.

LEDs are classified as solid-state lamps that have a few to 150 lm/W and their colour-rendering index is about 65-90 which comparable to natural light (Cole, Clayton, and Martin 2015; Loiselle, Butler, and Brady et al. 2015). Figure 4(a)depictsablock diagram of typical low wattage LED ballast. As LEDs are based on diodes that emit light when a dc voltage is applied (see Figure 4(a)), they require a constant current source from a low dc voltage source. Thereby, they are equipped with a very small-scale dc-dc converter to regulate the voltage and current fed to the LED (Oliveira et al. 2007). The buck, boost, fly back and resonant converters are normally used as a power source to the LEDs (Oliveira et al. 2007).The input current can be varied by the use of a triac-based dimmer circuit to vary the light output. Circuit complexity, step-down capability and line side converter spectral performance are the main considerations in designing the ballasts of LEDs (Sichirollo, Alonso, and Spiazzi 2015).

On the other hand, CFLs have a colour-rendering index of more than 75 ('Compact fluorescent lamp' 2017). Figure 4(b) depicts a block diagram of a typical CFL ballast. The lamp is resistive, but the electronic ballast connected between ac voltage and the lamp for controlling the lamp current is a capacitive load ('How compact fluorescent...' 2009). The CFL ballast circuit is a single phase capacitor-filtered uncontrolled ac/dc converter (Yong et al. 2010). During pre-ignition, the resonant tank is a series-LC circuit with a high Q-factor. After ignition and during running the tank is a series-L, parallel-RC circuit with a Q-factor somewhere between high and low value. CFL requires a high voltage for ignition, and a current to preheat the CFL filaments. Its electronic ballast circuit first converts the ac input voltage to a dc voltage through a full-wave rectifier, which is then converted to an ac square-wave voltage which becomes a sinusoidal current and voltage using a resonant tank circuit (see Figure 4(b)). When the CFL is turned on, the lamp filaments are preheated, voltage and current increase and frequency decrease. The frequency decreases continuously until the voltage exceeds the CFL ignition threshold voltage and the lamp turns on. Once the lamp is turned on, the ac voltage, current, and frequency come to the normal level (Cunill, Sainz, and Mesas 2013; 'How compact fluorescent ...' 2009). The CFL's non-linear current can be reduced by some compensation techniques such as the method of (Aizawa 2010).

Halogen lamps, similar to standard incandescent lamps, have a tungsten filament covered by halogen gas in the bulb. When an ac voltage is applied to the lamp terminals, the filament begins to radiate light in which its density depends on the level of current passing through it ('The Halogen Lamp' 2017).

These lamps, due to their nature, draw currents with different shapes. The current drawn by a standard incandescent lamp and halogen one is pure sinusoidal (see Figure 5(a)) while the current drawn by CFLs and LEDs is distorted (see Figure 5(b-c)) in the presence of a sinusoidal voltage (see Figure 5(d)). It is to be noted that the CFLs have larger settling time versus halogen and LEDs due to their internal characteristic, as is evident from their active power consumption (see Figure 5(e)).

These lamps are compared in Table 1 from different perspectives including the electrical load characteristic, efficiency and the heat and C[O.sub.2] emissions.

3. Considered utility and consumer-oriented assessment criteria

The studied lamps are evaluated from utility and consumer-oriented aspects. The considered utility-oriented aspects are mainly the power quality and non-active power while the considered consumer-oriented parameters are their luminous efficacy, cost and lifespan.

Power quality is an important factor in assessing the quality of electricity supplied to customers as it is directly related to the cause of mal-operation and malfunction of utility and customers' equipment. Utilities have reports of domestic customers complaining about the burning and failing of their household devices such as refrigerators, TVs, and air conditioners due to the poor quality of their supplied power. The utilities may also be disadvantaged due to the increased power losses in lines and distribution transformers. As is evident from Figure 5(b-c), CFLs and LEDs draw distorted currents, which can be quantified as harmonic current injection magnitudes or THD. Current harmonics and THD are classified under power quality and altogether stand for 22% of the observed power quality problems for American customers according to (Bhattacharyya and Cobben 2011) (see Figure 6). This study aims to assess and quantify these two sets of criteria for the majority of lightings sold on the Australian market nowadays. It also aims to validate the compliance of the power quality criteria with the limits specified in the Electromagnetic compatibility - Limits for harmonic current emissions (equipment input current less than 16 A per phase) (IEC 61000 - 3 - 2, 2009) and its equivalent Australian/New Zealand and European standards (AS/NZS 61000-3-2, 2003; EN61000-3-2, 2006). According to these standards, the harmonic current limits are categorised based on the power consumption of that equipment (i.e. light bulbs in this study) as given in Table 2. Alternatively, based on IEC61000-3-2(2009), the 3rd and 5th harmonic currents should not exceed 86 and 61% of the fundamental, respectively.
Figure 6. Different power quality problems observed in the USA.

Harmonic              22%
Voltage Sag/Swell,    48%
Other,                 4%
Load Interaction,      5%
Capacitor Switching,   6%
Grounding,            15%

Note: Table made from pie chart.

Another interesting technical criterion which is investigated in this research is the comparison of the ratio of the fundamental and non-fundamental components of the apparent power for different kinds of lamps. Assuming, S and [S.sub.1], respectively, as the total apparent power of the lamp and its fundamental component (at 50 Hz), [S.sub.N] is the non-fundamental component of the apparent power, expressed as (IEEE Std-1459, 2010):

[S.sub.N] = [square root of [S.sup.2] - [S.sup.2.sub.1]] (1)

The other interesting assessed technical criterion is the active power consumption of each light bulb and its deviation from the rated power given on the packaging. The research also aims to determine the observed true power factor, based on (IEEE Std-1459, 2010), that these light bulbs impose to their power system, which directly increases the level of reactive current and harmonics drawn from the residential feeder. The study later targets to determine and analyse the turning on time of these lamps and their light settlement time (also referred to as light warming time).

4. Methodology

Initially, a detailed list of all residential lightings sold on the Australian market was prepared from the major distributors and supermarkets (i.e. Bunnings Warehouse[R] as well as Coles[R] and Woolworths[R] supermarkets) including their online stores, and they were purchased. This study is focused on all lamps that were available on the market at the conduction stage ofthe project that includes 34 LEDs, 18 CFLs, and 21 halogen lamps of different brands, with different ratings and light colours (warm white and cool daylight). Table 3 lists the details of these lamps. The considered consumer-oriented parameters are compared using the data provided on the packaging of these lamps. The technical parameters are compared using practical laboratory-based measurements (see Figure 7) using a Fluke[R] power analyser (model 435 series II) and the Power Log 430-II (version 5.2) software that are used to record the power, voltage and current harmonic spectra of the lamps. All measurements have been recorded when the power consumption of the lamps have been stabilised (i.e. the data has been recorded 10 min after turning on the lamps). The data was recorded at one-second intervals, and the measurements were taken for a period of ten minutes for each lamp so that the lamp will be at the stable operation mode. The captured data was then analysed in MATLAB[R] to conduct a multi-criteria assessment of lamps based on different perspectives including harmonic current injection, THD, power factor, active power consumption, and stabilisation time, fundamental and non-fundamental apparent power and were then compared to each other to yield a better understanding of their characteristics.

5. Evaluation of consumer-oriented criteria

The studied lamps (indicated in Table 3) are evaluated based on their luminous efficacy, cost, and life-span in this section.

5.1. Luminous efficacy

Figure 8 illustrates a detailed comparison between the given illuminations of the studied lamps versus their rated active power consumption (both using the data available on their packaging). This is calculated as lumens per watt (lm/W) and is illustrated for different types oflamps separately. From Figure 8(a), it can be seen that there is a considerable difference between the maximum and minimum lm/W of LEDs, CFLs, and Halogen lamps. The average lm/W of Halogen lamps is 15.24 while this figure for CFLs and LEDs is 62.37 and 92.72, respectively, which is shown in Figure 8(b-d). Even the minimum lm/W of LEDs is larger than the maximum lm/W of CFLs. The same relationship is true between CFLs and halogen lamps as the minimum lm/W of CFLs is larger than the maximum lm/W of halogens. Figure 8(b-d) illustrates the variations of lm/W for different brands of the same lamp type. Among LEDs, it can be seen that on average, the Philips LEDs offer the maximum lm/W whereas the Crompton LEDs offer the minimum lm/W. In case of CFLs, the Coles CFLs offer maximum lm/W whereas the Philips CFLs offer the minimum lm/W whereas for halogen lamps, the Osram and Coles brand offer the maximum lm/W while the Philips and Nelson brand offer the minimum lm/W.

The study also shows that for the majority of studied lamps (except those of Click and Osram), the lm/W increases with an increase in the rated power of the lamps. It is also interesting to report that this change is less for CFLs than that of LEDs.

Figure 8(e) demonstrates the lm/W variation for two different brands (Philips and Osram) for two different colours of warm white (denoted by WW), and cool daylight (denoted by CDL) for their available CFLs and LEDs (given in Table 3). From this figure, it can be seen that LEDs with cool daylight colour have better lm/W compared to those with warm white colour. However, CFLs with warm whitecolourhavelargerlm/Wversusthosewith cool daylight colour.

5.2. Purchasing cost

Figure 9 illustrates a detailed comparison between the cost of the studied lamps versus their rated active power consumption, which is calculated and presented as Australian cents per watt (c/W). Similar to Figure 8,this figure illustrates the c/W of different types of lamps in separate subfigures. Figure 9(a) shows that halogen lamps are the cheapest ones in the market with an average of 5.63 c/W while CFLs and LEDs are more expensive with an average of, respectively, 46.4 and 120.04 c/W. Even the cheapest CFL (24c/W) is more expensive than the most expensive halogen lamp (9.79c/W). Figure 9(b-d) illustrates the variations of c/W for different brands of the same lamp. It canbeseenthatonaverage,OsramLEDsarethe most expensive (140 c/W) whereas Crompton is the cheapest (67 c/W). In case of CFLs, Philips has the most expensive (54 c/W) CFLs whereas Osram is the cheapest (25 c/W) whereas for halogen lamps, Philips is the most expensive (8 c/W) while Brilliant is the cheapest (3 c/W).

5.3. Lifespan

AllstudiedLEDshavealifespanof15thousand hours according to their packaging (see Figure 10). In case of CFLs, the lamps of Brilliant brand have a lifespan of 6 thousand hours whereas this figure is 10 thousand for the Osram and Coles CFLs. This figure is one thousand for the Philips halogen lamps, and 2 thousand for all other brands of halogen lamps.

6. Evaluation of utility-oriented criteria

This section analyses the potential power quality issues, as well as apparent power and power factor, in addition to the power consumption for the selected lamps.

6.1. Harmonic injection

The study shows that halogen lamps have a negligible harmonic injection, as seen from Figure 11(a). In this figure, the variations of each harmonic for the analysed halogen lamps is illustrated using a box plot which illustrates the maximum, minimum, average and distribution of the measurements for that harmonic level. In addition, the upper and lower limits of all measurements are further plotted. Further analysis reveals that this type of lamps have higher components in harmonics 5, 7 and 9, which consist 71% of all the injected harmonics (see Figure 11(a)).

The study also shows that LEDs and CFLs have a much larger current harmonic injection as compared to halogen lamps. Figure 11(b) illustrates the variations of the measurements of each harmonic component of LEDs and CFLs. From this figure, it can be seen that the average of each harmonic order injected by LEDs is larger than that of CFLs; however, this difference does not exceed by 16.35% (which is seen for the 13th harmonic). This figure also shows that the observed minimum harmonic values for LEDs are much lower than those of CFLs.

The experiments validate the current harmonics injected by all analysed LEDs and CFLs except one (i.e. the 10W Osram LED) are much above or slightly above the acceptable limits of Table 2 (IEC 61000-3-2; 2009 and AS/NZS 61000.3.2; 2003). Referring to the alternative technique given in (IEC 61000-3-2; 2009), the 3rd current harmonic generated by all LEDs and CFLs is less than the 86% limit. The 5th current harmonic generated by almost 40% of LEDs and 100% of all CFLs is less than the 61% limit.

6.2. Current THD

The experimental results show that as the halogen lamps behave as linear loads and inject very little harmonics into the supply system, their current THD is very minimal (not more than 1.97% on average for the studied lamps). On the other hand, CFLs and LEDs, being non-linear loads, injecting a quite large amount of current harmonics into the supply system, have a considerable current THD. The studies show that the average THD of CFLs and LEDs are, respectively, 102.61% and 125.7% (see Figure 12(a)). Figure 12(b-c) illustrates a comparative analysis of current THD of different brands of CFLs and LEDs, respectively. The study shows that 81% of CFLs have a current THD below 105% while this figure is only 9% for different brands of LEDs.

6.3. Power consumption

The results of experiments demonstrate that the level of the non-fundamental component of the apparent power of the studied LEDs and CFLs is higher than the fundamental component. This difference is more for LEDs and their average of SN/S is about 77% while their average of [S.sub.1]/S is almost 62% (see Figure 13). From this figure, it can be seen that the content of the non-fundamental component of the current of these lamps is more than that of their fundamental component. In case of halogen lamps, as they are largely linear loads, the content of the non-fundamental component of their apparent power is negligible (less than 3%).

The studies also show that on average all analysed lamps consume more active power than their rated power (given on their packaging). Figure 14 (a) illustrates a comparative analysis of this difference in the power consumption for LEDs, CFLs and halogen lamps separately while this difference is demonstrated for different brands of each lamp in Figure 14(b-d). It can be seen that on average, the halogen lamps consume more power (between 2-6% more) than their rated power (see Figure 14 (b)).Likewise, the range of active power deviation for LEDs is almost [+ or -]5% whereas it is about [+ or -] 10% for CFLs. It is worth mentioning that the analysis reveals that only 22% of CFLs and 15% of LEDs consume exactly the same power given on their packaging.

6.4. Settling time of lamps

The conducted experiments confirm that in overall, LEDs stabilise much faster than CFLs after turning on. Figure 15(a-b) demonstrates the stabilisation time of LEDs and CFLs for time slots of 0-10, 10-100, ..., 500-600 s. It can be seen from these figures that almost a quarter of the studied LEDs stabilise in less than 10 s, while none of the available CFLs stabilise in this period. Also, it can be seen that almost half of the LEDs stabilise in less than 200 s whereas only a quarter of the available CFLs stabilise in this time. This is a clear advantage of LEDs over CFLs. The variation range of the stabilisation time is also illustrated for different brands of LEDs and CFLs are shown, respectively, in Figure 15c-d). It is interesting to see that LEDs manufactured by Olsent have minimum average stabilisation time versus other brands while their CFLs have the longest stabilisation time among other CFL manufacturers. This figure is almost the same for both CFLs and LEDs manufactured by Philips.

6.5. Power factor

The studies show that on average, CFLs have a better power factor, measured based on (IEEE Std-1459, 2010), than LEDs but the difference is not quite significant (less than 5% approximately), as seen from Figure 16. It is to be noted that, as expected from halogen lamps, their power factor is an absolute unity.

Table 4 illustrates a comparison between the power factor and current THD of LEDs and CFLs with the same rated power. From this table, it can be seen that LEDs with low THD have high power factor and vice versa. This is true for the majority of the studied LEDs (except 6W Mirabella, Crompton, and Philips). However, this is not valid for the majority of CFLs. These exceptions are due to the fact that power factor is influenced not only by harmonic component but also by the reactive power consumption of each lamp.

7. Multi-criteria assessment

To compare different types of lamps from different brands based on the introduced utility and consumer-oriented criteria, a multi-criteria assessment is presented here. The considered utility-oriented parameters are the 3rd and 5th injected current harmonic, current THD, [S.sub.1]/S and [S.sub.N]/S, power factor while the consumer-oriented parameters are their illumination and cost in lm/W and c/W, as well as the lifespan. The results of all studied lamps are illustrated in Figure 17 in the form of radar charts in normalised values. This figure shows that 91% of the studied LEDs have a very similar utility and consumer-oriented characteristics and only 3 LEDs (i.e. the 10W Osram, 6W Mirabella, and 5W HPM) have some sort of distinguished differences from the rest. It can also be seen that the 10W Osram LED has the most superior utility-oriented performance; however, it is also the most expensive one on the market. In case of CFLs, the 5W Philips seems to be quite expensive while it nearly shows the same performance to the rest of CFLs from the perspective of other criteria. All of the studied halogen lamps also show a very similar behaviour on the basis of analysed parameters.

The study reveals that the 18W Philips and the 4.7W Osram LEDs have, respectively, the maximum and minimum lm/W (of 111 and 74). In the case of CFLs, the 20W Olsent and 5W

Philips, respectively, have the maximum and minimum lm/W (of 67 and 57) while the 72W Coles and 28W Philips halogen lamps have, respectively, the maximum and minimum lm/W (of 19 and 12). Also, the 10W Osram and 9W Click LEDs are, respectively, the most expensive and cheapest ones (with a cost of 200 and 56 c/W). In the case of CFLs, the 5W Philips and 15W Osram are, respectively, the most expensive and cheapest lamps (with a cost of 130 and 24 c/W) while the 28W Philips and 70W Brilliant are the most expensive and cheapest halogen lamps, respectively (with a cost of 10 and 2 c/W).

From the current THD perspective, the 10W Osram, 6W Mirabella, and 5W HPM LEDs are found to have minimum current THDs (respectively, 13, 57, and 65%) while this figure is above 113% for the other studied LEDs. Such low THDs reveal that some types of efficient filtering circuits have been used in these LEDs. On the other hand, the 4.7W Osram LED was found to have the largest current THD (of 153.03%). In the case of CFLs, the 15W Osram and 20W Brilliant have, respectively, the least and most current THD (of 97 and 112%, respectively).

From the power factor perceptive, the 10W Osram and 5W Click LEDs have the maximum and minimum power factors (of, respectively, 0.94 and 0.5). In the case of CFLs, the 24W Philips and the 20W Brilliant are found to have the maximum and minimum power factor (of, respectively, 0.64 and 0.59).

From the active power consumption perceptive, the maximum deviation of -15% for the 6W Mirabella LED, +14% for the 5W Philips CFL, and +7% for the 42W Philips halogen lamp were observed. It is also found that the 4.5W Osram and the 9W Click LEDs have the shortest and longest stabilisation times (of, respectively, 0 and 530 s). In the case of CFLs, the 5W Philips and the 14W Olsent have, respectively, the shortest and longest stabilisation times (of 11 and 572 s).

From the S1/S and SN/S perspectives, the 10 and 4.5W Osram LEDs have, respectively, the maximum and minimum S1/S (of 99 and 52%). These two lamps have, respectively, the smallest and largest SN/S (of 14 and 85%). For CFLs, the 15W Osram and 20W Brilliant have, respectively, the highest and lowest S1/S (of 0.72 and 0.67) and thereby the minimum and maximum SN/S (of 70 and 75%).

8. Conclusion

With the increasing awareness of consumers in using energy-efficient lamps, LEDs and CFLs are taking a larger share of the market versus halogen lamps. This study presented a multi-criteria assessment of the different residential lighting bulbs available on the Australian market. From the consumer-oriented criteria, the study shows that the illumination to power consumption ratio of LEDs is better than those of CFLs and this figure is far better than those of halogen lamps. This efficiency has been a strong motivation for Australian Building Code in developing regulations that promote CFLs and LEDs and bans energy inefficient lamps such as halogen ones. A comparative analysis shows that LEDs with cool daylight colour have better illumination than those with warm white colour, but this is opposite for CFLs. Comparing the cost of lamps versus their power consumption, it is found that LEDs are the most expensive lamps while halogen ones are the cheapest. There is a significant difference between the average cost of the LEDs, CFLs and halogen lamps, which may result in a different selection preference amongst consumers depending on their budget.

From the utility-oriented criteria, and through laboratory-based measurements, it is seen that all analysed CFLs have an acceptable 3rd and 5th current harmonic injection. In the case of LEDs, all of them have an acceptable 3rd harmonic but 60% of them have non-acceptable 5th current harmonic. However, they do not comply with the harmonic levels defined in IEC, European, and Australian standards for electrical loads of smaller than 25W. Overall, 83% of CFLs have a current THD of less than 105% while this figure is only 9% for LEDs. It is also found that majority of LEDs with high current THD have low power factors. On average, CFLs showed better power factors than LEDs. The measurements also illustrate that the level of non-fundamental component ofthe apparent power of LEDs and CFLs is higher than their fundamental component. It is also observed that only 22% of CFLs and 15% of LEDs consume exactly the same active power as given on their packaging. It is also found that in overall, LEDs stabilise much faster than CFLs after turning on, which is a clear advantage for LEDs.

The conducted multi-criteria assessment based on both utility and consumer-oriented criteria illustrates that all halogen lamps have a very similar performance from all considered criteria and this is true for all CFLs available on the market except one and for all LEDs except three. The largest difference among studied CFLs and LEDs is from the cost perspective, which is also linked to their luminous efficacy difference while the variation of their utility-oriented parameters is not very significant.

It is noteworthy that, this research was mainly focused on the power quality, reactive power, fundamental and non-fundamental power components and stabilisation time of the residential light bulbs available on the Australian market. To further generalise the comparison, they should be also evaluated from light spectral distribution and safety perspectives, which can be considered as a future research avenue.

Disclosure statement

No potential conflict of interest was reported by the author.


Farhad Shahnia [iD]


"How compact fluorescent lamps work and how to dim them," Website, 2009, Retrieved on 30 March 2017.

Aizawa, T. 2010. "Reduction of Neutral Conductor Current in a Three-Phase Four-Wire Low-Voltage Distribution Line." Electrical Engineering in Japan 171, (4): 19-27. doi:10.1002/eej.20897.

Arefi, A., J. Olamaei, A. Yavartalab, and H. Keshtkar. 2012. "Loss Reduction Experiences In Electric Power Distribution Companies of Iran." Energy Procedia 14,: 1392-1397. doi:10.1016/j.egypro.2011.12.1107.

AS/NZS 61000.3.2. 2003. Electromagnetic compatibility (EMC), Part 3.2: Limits- Limits for harmonic current emissions (equipment input current less than or equal to 16 A per phase). Australian/New Zealand Standard.

BCA Lighting Restrictions, Website, Retrieved on 30 March 2017.

Bessho, M., and K. Shimizu. 2012. "Latest Trends in LED Lighting." Electronics and Communications in Japan 95, (1): 1-7. doi:10.1002/ecj.10394.

Bhattacharyya, S., and S. Cobben. 2011. "Consequences of poor power quality- An overview." In Power Quality, Ed. A. Eberhard. InTech.

Bravo, R. J., and N. Y. Abed. 2013. "Experimental Evaluation of the Harmonic Behaviour of LED Light Bulb." IEEE Power & Energy Society General Meeting, Vancouver 1-4.

Cole, M., H. Clayton, and K. Martin. 2015. "Solid-State Lighting: The new normal in lighting." IEEE Trans. Industry Applications 51, (1): 109-119. doi:10.1109/TIA.2014.2328790.

Compact fluorescent lamp, Website, Retrieved 4 May 2017.

Comparison chart: LED lights vs. incandescent light bulbs vs. CFLs, Website, Retrieved on 13 March 2017.

Cuk, V., J. F. G. Cobben, W. L. Kling, and R. B. Timens. 2010. "An analysis of diversity factors applied to harmonic emission limits for energy saving lamps," 14th Int. Conf. on Harmonics and Quality of Power (ICHQP), pp.1-6, Bergamo.

Cunill, J., L. Sainz, and J. J. Mesas. 2013. "Neutral conductor current in three-phase networks with compact fluorescent lamps." Electric Power Systems Research 103,: 70-77. doi:10.1016/j.epsr.2013.05.006.

Duke, J. 2016. "Record number of new homes built in Sydney in 16 years," Website, Retrieved on 30 March 2017.

EN 61000-3-2. 2006. Electromagnetic compatibility (EMC), Part 3-2: Limits for harmonic current emissions (equipment input current [less than or equal to] 16A per phase). European Standard.

"Energy efficient lighting in houses, townhouses and units: For new dwellings and home renovations," Queensland Government, Department of Housing and Public Works, Information Sheet, 2013, Retrieved on 30 March 2017.

Energy savings forecast of solid-state lighting in general illumination applications, Office of energy efficiency and renewable energy, US Department of Energy, 2016. Retrieved on 30 March 2017.

"Energy use in the Australian residential sector 1986-2020," Australian Department of the Environment, Water, Heritage and the Arts, 2008, Retrieved on 30 March 2017.

Fuchs, E. F., and M. A. S. Masoum. 2008. Power Quality in Power Systems and Electrical Machines. Academic Press.

Gil-de-Castro, A., S. K. Ronnberg, M. H. J. Bollen, and A. Moreno-Munoz. 2014. "Study on harmonic emission of domestic equipment combined with different types of lighting." International Journal of Electrical Power & Energy Systems 55,: 116-127. doi:10.1016/j.ijepes.2013.09.001.

The Halogen Lamp, Edison Tech Center Website, Retrieved on 21 February 2017.

Hennessey, A. 2015. "Building approvals for new WA homes hits record high for second year," Website, Retrieved on 30 March 2017.

"Household energy efficiency Fact Sheet," Australian Clean Energy Council, Retrieved on 30 March 2017.

"How much electricity is used for lighting in the United States?" US Department of Energy market studies on lighting, 2016, Retrieved on 30 March 2017.

IEC 61000-3-2. 2009. Electromagnetic compatibility (EMC), Part 3-2: Limits- Limits for harmonic current emissions (equipment input current <16 A perphase). IEC Standard.

IEEE Std-1459. 2010. Definitions for the measurement of electric quantities under sinusoidal, non-sinusoidal, balanced or un-balanced conditions. IEEE Standard.

Johanson, S., "Australian homes still the world's biggest," The Sydney Morning Herald, Published on 22/ 08/2011, Retrieved on 30 March 2017. gest-20110821-1j5ev.html

Kooroshy, J., A. Ibbotson, B. Lee, D. R. Bingham, and W. Simons, "The low carbon economy: Equity investor's guide to a low carbon world 2015-25," Technical Report, Goldman Sachs, 2015. Retrieved on 26 May 2017.

Lam, J. C. W., and P. K. Jain. 2008. "A modified valley fill electronic ballast having a current source resonant inverter with improved line-current total harmonic distortion (THD), high power factor, and low lamp crest factor." IEEE Transactions on Industrial Electronics 55 (3): 11471159. doi:10.1109/TIE.2007.907668.

LEDs to account for quarter of lighting market by 2016, Website, Published on 19/ 04/2013,Retrieved on 4 May 2017.

Liang, X. 2017. "Emerging power quality challenges due to integration of renewable energy sources." IEEE Trans. Industry Applications 53, (2): 855-866. doi:10.1109/TIA.2016.2626253.

Lighting Catalog: Lamp Specification Guide, Philips, 2013. Retrieved on 30 March 2017.

Loiselle, R., J. Butler, G. Brady, et al. 2015. "LED lighting for oil and gas facilities." IEEE Trans. Industry Applications 51 (2): 1369-1374. DOI:10.1109/TIA.2014.2365357.

Matvoz, D., and M. Maksic. 2008. "Impact of compact fluorescent lamps on the electric power network," 13th Int. Conf. on Harmonics and Quality of Power (ICHQP), pp.1-6, Australia.

Meyer, J., A. M. Blanco, M. Domagk, and P. Schegner. 2017. "Assessment of prevailing harmonic current emission in public low-voltage networks." IEEE Trans. Power Delivery 32, (2): 962-970. doi:10.1109/TPWRD.2016.2558187.

Molina, J., J. J. Mesas, and L. Sainz. 2014. "Parameter estimation procedure for the equivalent circuit model of compact fluorescent lamps." Electric Power Systems Research 116,: 128-135. doi:10.1016/j.epsr.2014.05.009.

Oliveira, A. A. M., T. B. Marchesan, R. N. Prado, and A. Campos. 2007. "Distributed emergency lighting system LEDs driven by two integrated flyback converters." IEEE Industry Applications Society Annual Meeting, 1141-1146. New Orleans.

Phase out of incandescent light bulbs, Website, Retrieved on 15 May 2017.

Sichirollo, F., J. M. Alonso, and G. Spiazzi. 2015. "Anoveldouble integrated buck offline power supply for solid-state lighting applications." IEEE Transactions on Industry Applications 51 (2): 1268-1276. doi:10.1109/TIA.2014.2350071.

Sorensen, E., "Why are our houses getting bigger?" Website, 2013, Retrieved on 30March 2017.

Watson, N. R., T. L. Scott, and S. J. J. Hirsch. 2009. "Implications for distribution networks ofhigh penetration of compact fluorescent lamps." IEEE Trans. Power Delivery 24, (3): 1521-1528. doi:10.1109/TPWRD.2009.2014036.

Wells, Q. 2013. Smart Grid Home. Cengage Learning.

Window into Housing 2015 (Facts about the housing and construction industry in Australia), Housing Industry in Australia, Fact Sheet. Accessed on 30 March 2017.

Yong, J., L. Chen, A. B. Nassif, and W. Xu. 2010. "A Frequency-domain harmonic model for compact fluorescent lamps." IEEE Trans. Power Delivery 25, (2): 1182-1189. doi:10.1109/TPWRD.2009.2032915.

Muhammad Usman, Farhad Shahnia [iD], GM Shafiullah and Ali Arefi

School of Engineering and Information Technology, Murdoch University, Perth, Australia

CONTACT Farhad Shahnia [??] [??] School of Engineering and Information Technology, Murdoch University, Perth, Australia


Residential lamps; energy efficiency; multi-criteria assessment; power quality


Received 13 February 2018

Accepted 1 July 2018
Table 1. Comparison of characteristics of different types of
residential lightings.

                               LED         CFL         Halogen

Electrical characteristic      Non-linear  Non-linear  Linear
Turning on time                Instantly   Few         Instantly
Energy efficiency (Bravo and   3-4 times   1.75 times  Very
Abed 2013)                     of          of          small
                               halogen     halogen     (5%
                               lamps       lamps       light)
Heat emission (btu/hr)           3.4         30           85
("Comparison chart" 2017)
C[O.sub.2] emission (lb/year)  451         1,051       4,500
("Comparison chart" 2017)
Colour-rendering index          65~90        >75         100
(Loiselle, Butler, and Brady
et al. 2015; Oliveira et al.
2007; Sichirollo, Alonso, and
Spiazzi 2015;("Compact
fluorescent lamp" 2017)
Toxic Mercury                              [check]
Different colours              [check]     [check]
Shock resistance               [check]
Dimmable                       [check]     [check]

Table 2. Maximum harmonic current injection limit for light bulbs,
categorised based on their power consumption (IEC 61000-3-2; AS/NZS
61000.3.2:2003; EN 61000-3-2:2006).

                Maximum Harmonic
                % of the fundamental
Harmonic Order  component             mA/W

 2               2
 3              30 x power factor      3.4
 5              10                     1.9
 7               7                     1
 9               5                     0.5
11               3                     0.35
Between 13 and   3                     3.85 / harmonic
39                                     order

Table 3. List of available residential lightings on the Australian
market (analysed in this research).

Brand                       LEDs  CFLS  Halogen Lamps

Brilliant                          1     3
Click                        3
Coles                        1     2     3
Crompton                     1
HPM                          1
Mirabella                    3           3
Nelson                                   2
Olsent                       2     2     2
Osram      WW (*)            7     2     5
           CDL (#)           6
Philips    WW (*)            4     5     3
           CDL (*)           3     6
           CD ([infinity])   3
Total                       34    18    21

(*) WW = warm white, (#) CDL = cool daylight, ([infinity]) CD = Classic

Table 4. Comparison of current THD and power factor for lamps with the
same rated power.

Lamp   Power                   Current THD  Power
Type   (W)    Brand            (%)          Factor

LEDs    5     HPM               65.55       0.56
              Olsent           123.06       0.52
              Click            125.16       0.50
        7.5   Philips          112.86       0.61
              Osram            129.10       0.60
        9     Philips          123.98       0.57
              Mirabella        124.94       0.57
              Click            129.75       0.55
       10     Osram             13.10       0.94
              Coles            129.98       0.59
       10.5   Osram            125.23       0.61
              Philips          144.93       0.54
CFLs   15     Osram             96.76       0.61
              Coles             99.65       0.60
              Philips CDL (*)  100.12       0.63
              WW (#)           102.35       0.63
       20     Osram             99.64       0.61
              Philips CDL (*)  103.00       0.63
              WW (#)           103.21       0.63
              Olsent           104.34       0.61
              Brilliant        112.01        0.59
((*) WW = warm white, (#) CDL = cool daylight)
COPYRIGHT 2018 The Institution of Engineers, Australia
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Usman, Muhammad; Shahnia, Farhad; Shafiullah, G.M.; Arefi, Ali
Publication:Australian Journal of Electrical & Electronics Engineering
Geographic Code:8AUST
Date:Mar 1, 2018
Previous Article:Applications of artificial intelligence in powerline communications in terms of noise detection and reduction: a review.
Next Article:Power loss minimisation of rural feeder of Jaipur city by renewable-based DG technologies.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters