1. A brief overview of compact fluorescent lamps
Compact fluorescent lamps (CFLs) appeared some 20 years ago with the aim of replacing incandescent (tungsten filament) lamps. They are a promising type of lamp that lasts far longer and uses significantly less energy than a tungsten lamp of the same light output. Today they are available on the market in two main types: those with integrated electronic starting and those that require an external ballast, either magnetic or electronic.
The first type was designed as a direct replacement for incandescent lamps. It is available with the common E27 screw cap or the B22 bayonet cap, as well as with smaller caps (E14). The discharge tubes are either exposed or enclosed in a glass envelope shaped like that of a tungsten lamp. The starting circuit is electronic and is built into the base. For this reason lamps of this type are usually called electronic lamps. They are also widely referred to as energy-saving lamps (energy savers).
The second type is intended almost exclusively for use in luminaires (usually downlights or table lamps), because of its particular cap and the need to connect it to an external ballast. It comes in several versions, differing mainly in the type of ballast required. Lamps with a 2-pin cap usually work only with a magnetic ballast and have the starter built into the base. They cannot be dimmed, which rules out their use with the building management systems (BMS) of “intelligent” buildings to save energy on lighting. Lamps with 4 pins work with an external electronic ballast and therefore have no starter in the base. Their light output can be controlled electronically, and they work with building management systems.
The efficacy of compact fluorescent lamps, that is, the luminous flux (in lumens) the lamp delivers per watt of electrical power consumed, lies in the range of 55-75 lm/W (the power of the ballast is not included). This is slightly lower than the efficacy of the familiar linear fluorescent lamps, which can approach 100 lm/W in widely available types. Manufacturers claim that a 75 W incandescent lamp can be replaced by a 15 W electronic lamp, while a 20 W electronic lamp gives the same light as a 100 W incandescent.
It should be noted that compact fluorescent lamps are sensitive to temperature and to their operating position. Installing them in luminaires that do not dissipate properly the heat that inevitably builds up inside will result in significantly lower efficacy. Another characteristic is that their output falls considerably over time, a trait that other lamps show to a lesser degree.
The small size of electronic lamps and the ease of installing them (especially in the domestic sector) have placed them at the forefront of various lighting energy-saving campaigns run by electricity companies and other organisations. Their currently rather high purchase price is often subsidised. The Public Power Corporation (PPC) has at times subsidised their installation on Aegean islands where the cost of generating electricity is very high, especially on islands with weak, stand-alone renewable energy systems (photovoltaics, etc.). These campaigns have led to a steadily growing demand. For example, only a few hundred electronic lamps had been sold in Greece by 1989, whereas in 1998 alone 100,000 such lamps were installed on a number of Aegean islands and in Thrace, according to the PPC’s annual report for that year.
For this type of lamp to gain wider acceptance among the general public, its advantages must be demonstrated with reliable arguments, so as to overcome the hesitation caused by its high purchase price. Unfortunately, the companies’ arguments to date have not delivered the expected results. Consumer distrust is partly justified, particularly in our country, which is marked by a lack of credibility. It takes a great deal of effort to convince consumers that, in order to help their country, they must themselves pay a considerable sum. What will win them over is the prospect of a real, personal benefit from the lamp’s high efficacy and long life. For that to happen, however, the lamp’s technical advantages must be certified by independent and reliable bodies, which has not yet been done comprehensively.
Consumers view the exaggerated performance figures printed on the packaging of these lamps with suspicion, especially given an inherent weakness that even the least informed consumer can notice. Their brightness drops noticeably within the first thousand hours of operation and keeps falling to such a degree that the quality of lighting in the space deteriorates, and consumers are often driven to replace them early, which cancels out the advantage of their long life.
2. Laboratory investigation
In recent years the Lighting Laboratory of the National Technical University of Athens began an effort to investigate the advantages and disadvantages of compact fluorescent lamps with electronic starting. As part of this work, an attempt was made to record the light output of various lamp types throughout their life. For this purpose three types were chosen from the three largest manufacturers. The types selected, rated at 15 W, 20 W and 23 W respectively, are the most widespread. According to the manufacturers’ catalogues, their luminous flux is 900 lm, 1,200 lm and 1,500 lm respectively. To minimise statistical error, a sufficient number of lamps of each type was used. It should be noted that the cost of this project, including the purchase of the lamps, was covered from the laboratory’s own funds, without seeking support from any third party.
The luminous flux of the lamps was measured after the first 100 hours needed for them to stabilise, and measurements continued at 1,000-hour intervals until they failed for good, which happened after 12,000 hours. A timer switched the lamps off for 15 minutes every 3 hours to let them rest and to simulate more faithfully the conditions in which they are meant to operate in consumers’ homes. A stabiliser kept the voltage at 230 V.
Alongside the luminous flux measurements, other characteristics of the lamps were measured, such as their power (active and reactive), power factor and current harmonic distortion. A strong harmonic distortion of the current was observed, exceeding 100%. This creates problems for the power quality of the electrical network, as well as safety problems (overheating of cables, arcing in electrical appliances, etc.). Among other things, it increases reactive consumption, so that the power factor barely exceeds 0.60. A low power factor cannot easily be corrected with a compensation circuit, which, moreover, is already part of the integrated electronic starting circuit. All the integrated circuit achieves is a correction of the cosine of the fundamental, that is, a reduction of the phase difference between the voltage and the fundamental component of the current. With harmonically distorted currents, however, a low power factor is not the same as the cosine. So electronic lamps are seen to consume reactive power even though their cosine is close to unity.
The results of the luminous flux measurements are summarised in the figure. Small deviations appeared between samples of the same type from the same manufacturer. More significant differences appeared between manufacturers. One manufacturer’s lamps initially showed the highest output, but over time it fell considerably, so that the products of the other two manufacturers overtook them. In particular, it was observed that one manufacturer’s lamps started less well than the others but were more stable over time. Because the aim of the research was not to compare manufacturers but to draw general conclusions about electronic lamp technology, the measurements for each lamp type were averaged, so that each curve in the figure represents all the samples of the same wattage from the three manufacturers.
In general, all the electronic lamps showed a significant drop in light output during testing. After the first 1,000 hours their output falls by about 7%, and at 5,000 hours of life they deliver only 80-85% of their initial output. Towards the end of their life it falls to just 68-75% of the initial value. These figures are not at all satisfactory compared with the incandescent lamp, whose output at the end of its life falls to 90% of nominal. Consumers therefore face a dilemma: put up with the reduced lighting and so lower their quality of life, or replace the lamp before it dies? In the latter case they will of course lose the advantage of the lamp’s long life, and their financial benefit will shrink considerably. The energy savings they achieved may even fail to offset the high purchase price of the electronic lamp, leaving the disappointed consumer to return to the energy-hungry but cheap competitor with the tungsten filament.
Numerical integration of the luminous flux measurements over the lamps’ life showed that the 20 W electronic lamp, not the 15 W one as the manufacturers claim, should be regarded as equivalent to the 75 W incandescent lamp. Likewise, the 23 W electronic lamp, not the 20 W one, should be regarded as equivalent to the 100 W incandescent lamp.
3. The economic perspective
In an effort to investigate the real benefits of replacing incandescent lamps with electronic ones, an economic analysis was carried out comparing the running cost of a lighting installation using 100 W incandescent lamps with that of the same installation using 23 W electronic lamps, which, as mentioned above, produce the same lighting. In other words, the cost of lighting was calculated over a period of 12,000 hours, the life of the electronic lamp. Over this period the first installation will need 12 incandescent lamps, because their nominal life is 1,000 hours, while the second will need only 1 electronic lamp. Given that a lamp burns for about 2,000 hours a year in real conditions, the 12,000 hours of operation correspond to 6 years.
The calculation model was relatively simple, but since going into detail is beyond the scope of this article, it will simply be noted that the figure of €0.07 was used as the cost per kWh (for domestic consumers) and 7% as the borrowing interest rate. The purchase costs of the lamps are €12.0 for the electronic lamp and €0.6 for the incandescent. The calculation took into account the purchase cost of the lamps, the cost of the electricity consumed and the current interest rate on capital over the 6-year period.
The running cost of the installation with 100 W incandescent lamps was calculated at €109.47 for the 6-year period, while for the same period the cost of the installation with the 23 W electronic lamp came to €41.00. The ratio of the cost of the incandescent installation to that of the same installation with an electronic lamp is therefore 1:2.67, not 1:5. This means that the money saved by the domestic consumer by using the electronic lamp reaches 63%, not 80% as is widely claimed.
A similar calculation was made for the 75 W incandescent lamp against the 20 W electronic lamp. The cost ratio was found to be 1:2.41, so the saving in money is 59%. The financial benefit for industrial consumers turns out to be even lower, owing to the charge for peak apparent power (kVA).
4. The energy benefit
Although the results of the measurements and calculations are less enthusiastic than the widely held, but undocumented, views, the answer to the question in the title of this article is yes. In Greece in 1998 the installed generating capacity was 10,296 MW and annual electricity production was 41,834 GWh. Household consumption accounted for 40% of this energy. Since lighting accounts for 8% of household consumption, it follows that household lighting requires 3.2% of the total energy the country produces. According to statistics from Greek and European organisations, home lighting is provided mainly by incandescent lamps (halogen lamps fall into this category). It is therefore estimated that incandescent lamps in homes consume 2.4% of total energy. If the commercial, industrial and agricultural sectors are included, the conclusion readily follows that incandescent lamps consume 3% of the total electricity generated in the country.
Assuming that 25% of incandescent lamps are replaced with equivalent electronic ones, electrical power demand would fall by 60 MW. To this benefit would be added a reduction in air pollution, given that 70% of the country’s electricity generation uses lignite as fuel. The strategy of replacing tungsten filament lamps will contribute positively to the country’s compliance with European directives and the Kyoto Protocol on reducing atmospheric pollutants.
For the time being, a higher rate of replacement is not recommended, because the problems of harmonic distortion, which the substantial load of electronic lamps would inject into the electrical network, have not yet been solved. Once this problem is resolved, the way will be open for large-scale replacement of energy-hungry incandescent lamps. Researchers at the Lighting Laboratory of the National Technical University of Athens are working in this direction in cooperation with institutions abroad.
5. Conclusion
New technologies should neither be demonised nor idealised. Both approaches do equal harm. What is needed is an objective approach and a thorough study of their advantages and disadvantages. Only with the positive conclusions of such a study, and with those alone, will we arrive at lasting and sustainable solutions.
