Download pdf - Effects of Power Quality on Manufacturing Costs in Textile Industry

Scientific Research and Essay Vol.4 (10), pp. 1085-1099, October 2009 Available online at http://www.academicjournals.org/SRE ISSN 1992-2248 © 2009 Academic Journals Full Length Research Paper

Effects of power quality on manufacturing costs in textile industry

F. Koçyi�it1, E. Yanıko�lu1, A. S. Yilmaz2* and M. Bayrak1

1Sakarya University, Electrical-Electronics Engineering Department, Sakarya, Turkey.

2Sutcu Imam University, Electrical-Electronics Engineering Department, Kahramanmaras, Turkey.

Accepted 15 July, 2009

This paper reports the effect of electrical power quality on textile industry. For this purpose, power quality measurements have been done for six months in two different sectors of textile industry. All parameters affecting power quality have been measured by using ION 7650 Power Analyzer according to the EN 50160 standard. Because textile industries have high technology machines including electronic control cards and driver controlled motors, poor power quality may damage the system and cause production failure. Measurements showed that the losses caused by electrical power quality were significantly high, being around 15% of the annual net profit of the textile industry. Key words: Power quality, power quality metering, textile industry.

INTRODUCTION In the world’s new economic system, in other words, the global economy, the conditions for existence have become quite difficult for establishments. The fast-growing Far East, becoming a manufacturing center, has significantly pulled down the profit rates of industrial establishments. The profit percentages which were represented by two-digit and even three-digit figures have decreased to single-digit figures due to excess supply and intense competition. Both the shrinkage in the market and the decreases in profitability have forced industrial establishments to manage their costs effectively (Sullivan et al., 1996).

In order to manage their costs, the companies have not only oriented towards qualified human resources but also have started to pay attention to quality energy supply, which is another fundamental factor in the sector. Manufacturing loss due to poor power quality is reflected on the income statements and balance sheets of these companies, as an increased cost due to inefficiency. Along with this, in a world of such intense global competition, the penal outcomes as well as the adversities which can lead to losing the customer or *Corresponding author. E-mail: [email protected].

customers as a result of the non-delivery of products, which cannot be manufactured because of a machine break-down due to energy failure or poor power quality, are the results which cannot be endured by today’s establishments. Due to such reasons, the customer-focused establishments pay attention to quality power sources including electrical power and increasing their quality (Sözen et al., 2007; Grupta et al., 2004; Davis et al., 2000; Nooij et al., 2007).

Besides other sectors, the influence of electrical power quality on the textile sector and textile establishments have not been sufficiently analyzed with studies until now, being quite superficial. There has not been any study requiring on-site measurements for more concrete data until now. This study presents data from measure-ments and monitoring of over 6 months at an integrated textile establishment in Bursa and Kahramanmara�, Turkey.

For this, the energy quality measurements at a textile establishment were made by ION 7650 power monitoring and analysis device and the poor power quality was assessed in accordance with EN 50160 standard; the damage from such poor quality of energy suffered by the establishment was also calculated. The findings and the impacts on weaving, yarn, etc. departments have been analyzed in detail (Koçyi�it et al., 2008).

1086 Sci. Res. Essays

Table 1. Categories and typical characteristics of power system electromagnetic phenomena. Categories Spectral content Duration Voltage magnitude 1. Transients a. Impulsive Nanosecond 5 ns rise < 50 ns Microsecond 1 µs rise 50 ns – 1 ms Milisecond 0.1 ms rise > 1 ms b. Oscillatory Low frequency < 5 kHz 0.3-50 ms 0 - 4 pu Medium frequency 5-500 kHz 20 µs 0 - 8 pu High frequency 0.5-5 MHz 5 µs 0 - 4 pu 2. Short duration variations a. Instantaneous Sag 0.5 - 30 cycles 0.1 - 0.9 pu Swell 0.5 - 30 cycles 1.1 - 1.8 pu b. Momentary Interruptions 0.5 - 30 cycles < 0.1 pu Sag 30 cycles – 3 s 0.1 - 0.9 pu Swell 30 cycles – 3 s 1.1 - 1.4 pu c. Temporary Interruptions 3 s - 1 min < 0.1 pu Sag 3 s - 1 min 0.1 - 0.9 pu Swell 3 s - 1 min 1.1 - 1.2 pu 3. Long duration variations Interruptions sustained > 1 min 0.0 pu Undervoltages > 1 min 0.8 pu Overvoltages > 1 min 1.1-1.2 pu 4. Voltage imbalance SS 0.5 - 2% 5. Waveform distortions DC offset SS 0 - 0.1%

Harmonics 0-100 th H SS 0 - 20% Interharmonics 0-6 kHz 0 - 2% Notching SS Noise Broadband SS 0 – 1% 6. Voltage fluctuations > 25 Hz Intermittent 0.1 - 7% 7. Power frequency variations < 10 s

SS : Steady state DEFINITION OF POWER QUALITY Power quality (PQ) is defined by (IEEE Std 1159.3 2003) as “set of parameters defining the properties of the power supply as delivered to the user in normal operating condi-tions in terms of continuity of supply and characteristics of voltage (symmetry, frequency, magnitude and waveform)”. Also, PQ deals with not only voltage quality

but also current quality. It is the combination of voltage and current quality. In practice, there are several types of PQ disturbances such as voltage sag/swell/interruptions, switching transients, flickers, harmonics, notches, etc. caused by faults, nonlinear loads and dynamic operating conditions (IEEE Std 1159, 1995; Alkan and Yilmaz, 2006). Table 1 illustrates categories and typical charac-teristics of power system electromagnetic phenomena

Kocyigit et al. 1087

Figure 1. Weaving machines (Dobby machines).

(EN Std 50160, 1999). EFFECTS OF PQ ON WEAVING-KNITTING Textile weaving covers a very wide range of products. Most of the garments, surface coverings of the sitting sets in our rooms, tulle and curtains, carpets, towels in the bathrooms, and even the airbags in our cars are of fabrics manufactured through textile weaving or knitting. Today’s weaving and knitting machines are high technology machines. They include tens of sensors, driver controlled ac and dc electrical machines, communi-cations cards, cpu units, touchscreens, ethernet cards and encoders. These equipments operate at different voltage values. If any interruption, in the machines at the weaving room occurs for any reason, it can take up to 4 h for the machinery park to return to the acceptable performance values. Thus, any short-term or long-term interruption means a direct manufacturing loss. These machines which are sensitive to sinusoidal form of electrical power can shut down if a voltage drop, voltage sag and voltage fluctuations occur. These machines include special encoders used for position determination and interruptions without any reason can lead to the break down of this device (values approximately $500) and its replacement means at least 2 h of interruption. Voltage sags can result in excess current drawings, which cause the engines operating at approximately 6 kW to draw higher currents, leading to the break-down of electronic cards driving such engines. Such cards cost approximately $2500 and their replacement may take minimum 2 h.

At a typical weaving establishment shown in Figure 1 (sample establishments), the average daily turnover of a weaving loom can be considered to be $2,400. Such turnover indicates that the value added to the economy in 1 h by one machine is $100. In our country, the average

number of the weaving machines for each establishment in corporate companies is between 100 and 1000. In such establishments, it can be simply calculated that even a 1 h interruption means a weaving manufacturing loss between $10,000 and $100,000. When such interruption occurs at an organized industrial zone with several textile plants, the loss will reach to more dramatic values.

High current and voltage harmonics and the other distortions can cause such machines to stop. Harmonics shorten the life of all electrical and electronic equipments in the machines. They cause break-downs, resulting in repair costs and manufacturing losses. They also shorten the depreciation periods of the machines, leading to increasing costs and thus losses (De Abreu and Emanuel, 2002). EFFECTS OF PQ ON DYEING PROCESSES Dyeing processes in the textile sector can be classified under main categories of fiber, yarn and fabric dyeing. Despite the differences in physical structures of such machines, they operate under similar principles as the materials they dye are the same. All fiber, yarn and fabrics are made of raw materials such as acrylic, polyes-ter, cotton or viscose. These machines include driver-controlled pumps, level gauges, calorimeters, liquid meters, PLC devices or cpu including control cards, touchscreen panels, actuator valves, proportional valves and several other electrical, electromechanical and electronic sensors and equipments.

Dyeing machinery equipments can be interrupted in most of the poor quality electrical energy situations. The amount of losses suffered depends on the duration of such interruptions. In situations of voltage drop and volt-age fluctuations, the machinery may activate the automatic protection and shut itself down. If the poor

1088 Sci. Res. Essays quality situation lasts for short periods (before the water inside the kier cools), the repair process called "reinforcement" is performed. This results in extension of the process time as well as water, energy, chemical and depreciation losses. It usually costs the half of the dyeing expenditures. If we assume that polyester is dyed in a dyeing kier of 1000 kg, it means that the loss per kilogram will be $0.5, making a total loss of $500. In cotton or viscose dyeing processes, the loss is doubled. If the process involves fabric dyeing for automotive sector, the process may have to be repeated all over again. This results in the loss being doubled. A worse situation is the loss of the yarn or fabric in the dyeing kier. In some of the interruptions, the material dyed may have to be scrapped due to the nature of the raw material or product.

If there is yarn in a kier of one ton, and for example if it is polyester, the loss will be between $2,000 and $4,000. And if it is fabric, the loss will reach to $4,000 - $20,000. There are minimum 10 dyeing kiers in corporate dyeing houses. If these are assumed to dye polyester, we can say that they can perform the dyeing process 60 times a day. If we assume that the average kier capacity is 500 kg, the daily dyeing capacity would be 30 tons. Given that there are hundreds of dyeing houses in our country, the loss can be estimated from the above figures. Our coun-try’s textile sector has witnessed a significant downfall in the recent years. With each year, several dyeing houses are closed. The turnover of a dyeing house performing yarn-dyeing in a one-ton kier would be $1500, its best profit being $300. As explained above, a single energy problem can lead to a loss amounting to 500 or 1000 or even $2,000. Therefore, the significance of electrical quality is obviously seen. High quality electrical energy has an important role in the sustainability of such companies in today’s commercial environment of low profitability. EFFECTS OF PQ ON FINISHING PROCESSES Finishing process means the last processes performed on the textile products and involves a wide range of machines. The mid-part of the stenter machines is the heated drying section which is also known as the oven. This section varies between 15 and 35 m depending on the establishment. The fabric is stabilized through heating by hot oil or natural gas. Two different processes can be carried out before the drying process. The first process is putting the fabric into a chemical-filled basin at the inlet of the machine. Here the fabric is processed with chemicals, as a result of which the fabric attains features such as delayed ignition, anti-germ and in the case of jeans fabric, prevention of leg turn-ups. In the second process, the said chemicals are only applied to one surface of the fabric in the covering unit, again at the inlet of the machine.

The total length of the stenter machine is between 30 and 60 m. We can take the average length as 50 m. From inlet to the outlet of the machine, there are about

fifty driver-controlled motors. Within the panel of the machine, which is more than 10 m long, there are PLC, drivers, hundreds of control elements, sensors most of which have modbus communication, touchscreen control panels, computer-controlled weft straignteners operating synchronously with the machine, automatic fabric accumulator, J-Box machine and several other equip-ments. The electrical control panel of such machines is longer than 10 m.

A problem at any component of this complicated line ends up with the shut down of the line. The fabric passes through the machine at a certain speed. The speed range is between 10 and 100 m/d. At this stage, the tem-perature at the oven is between 100 - 180°C. The speed and the temperature vary depending on the process and fabric types. With each halt of the line which is sensitive to poor quality energy, the fabric left inside the machine is deformed from the heat and is scrapped. The value of any fabric with an average length of 50 m is between $100 and $1,000. Re-manufacturing of such fabric would be even more costly. If special chemicals are used, their cost will also be added as a loss.

The shortening of the depreciation period due to poor power quality is even more significant in this machine. Any such line costs between 750.000 and $1.250.000, depending on the accessories. There are hundreds of finishing establishments operating with such machines in our country. In the meantime, the equipments breaking down due to harmonics and other poor power quality problems are counted as direct loss. When the equipments break down, the machine stops and the time required for detection of the break-down and repair is a significant amount of time. In such situations, the manufacturing losses will reach to higher values. Let us give the most optimistic scenario. Let us assume that the cost of a one-meter process in a finishing process is $1. The processing speed of the machine is 30 m/d. Under this scenario, the manufacturing loss will be 1800 m per h, with a cost of $1800. If the machine remains out of operation for 10 h, the loss will be $18000. In the real establishments, there have been some break-downs with 10 days of repair period (waiting for replacements, etc.). Such break-downs were witnessed in the establishments where the measurements took place for this study. MEASUREMENT METHODS AND STANDARDS In our country, energy quality is monitored in accordance with the “Regulation on supply sustainability, commercial and technical quality of electrical energy provided through the distribution system in the electricity market” which is a translation of EN 50160 standard.

Measurements regarding power quality should be carried out in accordance with this regulation. To achieve this, the required devices should both momentarily monitor energy flow and all the parameters related to quality, as well as record the problems (quality deviations). It is essential


Figure 2. ION Enterprise software interface. that such devices are internationally accredited, in other words, the measurements performed should be legally recognized and acceptable. The measurements have been carried out with ION 7650 device, performing three-phase and neutral measurement which is compliant to the requirements above. The six months measurements performed at two different establishments within this study, were fully conducted and planned in accordance with EN 50160 standard.

The measurements were recorded real-time on the computer using the ION enterprise software. The inter-face of the software is given below (Figure 2). The power quality and continuous event records can be reached from an effective menu.

With the privatization of energy production, transmis-sion and distribution, energy has begun to be considered as a meta that should be produced and distributed in accordance with certain standards and rules among the companies supplying the energy and the industrial organizations utilizing the energy. Companies or organi-zations producing and/or selling energy have become liable to supply the electrical power they have to generate within certain boundaries called electrical quality, in an uninterrupted and trouble-free manner. In the industry,

the energy quality (or poor quality) is monitored and reported in accordance with certain power quality stan-dards and regulations. MEASUREMENT RESULTS All parameters affecting the power quality were measured momentarily (by taking 1024 samples from one period) and the related parameters were recorded on the main computer through ethernet connection. Besides recording of all parameters, the deviations from power quality (according to EN 50160) were kept in a separate file. Since the poor power quality was recorded in transient measurements, irrelevant data did not occupy space on the computer. As time was also recorded for energy quality deviations, the manufacturing losses and the reasons for the damage could be approximately estimated.

In selection of the establishments, attention was paid to choose them from different parts of the country. Com-bined with the features of the industrial zones they were located, it was demonstrated that those two plants were operating on electrical energy of very different quality


Figure 3. Successive voltage sag occurrences recorded.

values. One of our establishments was located in an organized industrial zone in the Marmara Region. In this organized industrial zone, electricity is fed through double lines and in addition to this the same main busbar is fed by three high power natural gas plants. As a result of the measurements performed, it was demonstrated that of the two establishments this one was operating on higher electrical quality. Naturally, the effects of and material loss from poor quality were insignificant in this establishment. On the other hand, in the establishment located in the Eastern Region, there was another autoproducer in the main busbar of the distribution line from which the establishment was supplied electrical energy. As it will be explained in the following sections, there were several poor power quality incidents in this establishment. This resulted in excessive material burden for the establishment. As a result of the measurements performed, it was seen that the poor quality of electrical energy incidents explained in the second section were seen in the first establishment for several times. In the other establishment, the incidents mentioned occurred fewer times than the first establishment. When the in-cidents occurred, the corresponding problems in the plant were also recorded. On the basis of such records, the costs were calculated in cooperation with the planning and production departments of the establishments. The calculations were related to the direct costs and the indirect costs such as loss of reputation and customers

were not included. Voltage sag measurements and their cost During the six-month-measurements in the first establishment, there were 155 voltage sag occurrences recorded. In the second establishment, on the other hand, the number of voltage sag occurrences was six. The depth and duration of such occurrences determined the effects on the establishment. The six occurrences in the second establishment did not have any effects on the establishment. The Figure 3 shows an example of the measurements and the voltage sag records. The current variations at the time the voltage sag occurred are shown in Figure 4. The occurrences resulted in manufacturing losses and breakdowns in the first establishment. Voltage sags with greater depth and duration caused some machines to shut down as well as some breakdowns. Voltage sags were in succession at times and they sometimes resulted in power failures. Successive voltage sags could be readily monitored. The records could be transformed into a graphic with a simple option in the software. In the first establishment, some of the voltage sags resulted in general interruption, some caused partial interruption and some others led to breakdowns and combinations. The records and calculations revealed that the six-month-occurrences resulted in a cost of $110,000.


Figure 4. Current variations in the successive voltage sag occurrences recorded.

The details are shown in Table 2. Transient measurements and their costs As a result of the measurement process, 44 transient occurrences were recorded in the first establishment while there were no transient records in the second establishment. In measurement of the transient occurrences the device was set to the highest sensitivity and the occurrences were momentarily recorded during two periods. The calculated number of the points during the two periods was 2048. Thus, the time between two points reached to a very sensitive value, nearly 20 µs. Figure 5 shows the graphic of the transient moments. Figure 6 shows the three phase current wave forms at the moment of the transient occurrence and Figure 7 demonstrates the voltage harmonics at the moment of the occurrence.

When the figures are analyzed, it is seen from the current reaching to a five-fold higher value and the releases occurring on voltage curves that a high-power condenser group was activated. The fact that the current harmonicsreach to high frequency values and their amplitude exceeds 100%, shows the occurrence of a resonance. This led to problems such as the explosion or breakdown of the condensers in the compensation panel, the burning and melting of the contactor contacts, as well as the mel-ting of the connected cables. Here, the greatest hazard is the continuous fire risk.

The cost of the problems such as material and labor loss as a result of such occurrences in the first establish-

ment was calculated to be around $1000. No such occurrence or related cost was seen in the second establishment.

Harmonic measurements and their costs The current and voltage harmonics were continuously measured until the 63th harmonic. As shown in Figures 8 and 9 both current and voltage harmonics measured in the both establishments were over the limit values. The values in the graphic represent the average values, so the harmonics occurring in the establishment are above such values. It is seen that the voltage harmonics exceed 6% and the current harmonics exceed 20%. The current harmonics frequently reach to 30 - 35%. In addition to the problems created at the compensation panel of the esta-blishment, these harmonic components can obviously damage the other machines and fixings operated in the establish-ment. The current and voltage magnitudes including such high-value harmonics can damage the electronic cards and create negative results on the lives of the other electrical machinery. Replacing the compen-sation panel of the establishment with a harmonic-filtered compensa-tion panel is compulsory, given the resonance occur-rences. Voltage interruptions One of the most significant causes of the losses in the establishments is the power failures. When the machines


Table 2. Costs of voltage sag in establishment 1.

Duration Phase A Phase B Phase C Problems encountered Cost ($)

0,4190 51,382 52,054 51,171 Partial Interruption, breakdown 14,500


0,0690 89,742 89,961 89,546

0,1600 88,386 89,149 80,255

0,1400 75,291 88,744 74,863 Partial Interruption 2,000

0,3210 93,074 88,362 90,621

0,3190 85,790 86,499 94,818


0,6000 81,873 81,814 81,660


0,1690 5,921 3,927 6,336 Interruption, breakdown 18,500

0,1190 91,258 88,700 92,205



0,1810 87,257 88,277 79,876

0,9100 76,054 77,386 75,856 Breakdown 3,000

0,9610 88,260 89,317 92,986

0,8790 88,495 89,276 93,221 0,26 80,831 80,257 80,373

0,3190 84,683 88,350 89,196

0,9910 80,595 81,383 81,185

0,1790 92,087 89,654 92,771 0,3090 80,662 81,456 80,934 0.4410 82,763 85,879 87,233

1,01 76,666 77,671 77,203 Breakdown 2,500

0,2290 80,475 81,388 80,791

0,7910 80,301 79,627 80,725

0,7900 77,507 78,690 78,503 Breakdown 2,500

0,9200 76,215 77,481 76,436 Breakdown 2,000

0,9800 86,202 86,877 86,708



0,1390 91,471 82,469 74,690

0,1300 92,969 77,311 48,946 Breakdown 2,000

0,1590 84,499 85,148 84,973

0,2100 87,402 87,646 78,818

0,3390 86,149 86,881 86,436

0,3180 88,878 92,976 88,897

0,3900 84,774 85,409 85,834

0,7210 89,309 86,212 87,191

0,0500 87,875 88,261 87,811


0,24 81,387 81,394 81,809



Figure 5. Three phase voltage wave form at the moment of the transient occurrence.

Figure 6. Three phase current wave form (at the moment of the transient occurrence in Figure 6).


Figure 7.Voltage harmonics (at the moment of the occurrence in Figure 6).

Figure 8. Variations in the 6-month voltage and current harmonics of the first establishment.


Figure 9. Variations in the 6 -month voltage and current harmonics of the second establishment.

Table 3. Electrical failures in establishment 1.

Time label V1-Irpt durtn V2-Irpt durtn V3-Irpt durtn

02/05/2008@ 12:07:33.267 PM 5098.174 5098.174 5098.174 03/05/2008@ 04:47:16.841 AM 869.347 869.347 869.357 03/05/2008@ 06:25:49.112 AM 4222.712 4222.712 4222.722 23/05/2008@ 08:12:59.655 AM 7374.369 7374.369 7374.369 24/05/2008@ 05:06:33.902 AM 3519.939 3519.939 3519.939 26/06/2008@ 10:14:03.382 AM 426.742 426.742 426.742 14/08/2008@ 06:29:08.813 AM 1217.348 1217.348 1217.348 15/08/2008@ 08:49:43.032 AM 835.605 835.605 835.605 19/09/2008@ 04:25:07.430 AM 2602.133 2602.133 2602.133

Table 4. Electrical failures in establishment 2.

Time label V1-Irpt durtn V2-Irpt durtn V3-Irpt durtn [email protected] PM 2035.566 2035.496 2035.496 [email protected] AM 12548.032 12548.032 12548.032 [email protected] AM 146.782 146.793 146.772 [email protected] AM 398.137 398.137 398.147 [email protected] AM 497.88 497.88 497.9 [email protected] AM 2297.264 2297.274 2297.264 [email protected] AM 419.267 419.267 419.267 [email protected] AM 2107.798 2107.798 2107.798 [email protected] AM 1241.322 1241.301 1241.311 [email protected] PM 18.47 18.47 18.47

1096 Sci. Res. Essays stop due to failures, their re-activation takes some time which results in reduced efficiency and manufacturing losses. Secondly, when the power failures last for long periods, the machines cool down and after their re-activation, there is loss in efficiency until they reach to the previous high efficiency rates. When the failures are uncontrolled interruptions, the mechanical and the driving electrical and electronic equipments in high speed and high moment machines may break down, resulting in an additional cost. The interruptions in establishment 1 and 2 are shown in Tables 3 and 4. The first column shows the time label of the occurrence. In the following three columns, the interruption periods for each phase are given in seconds.

In the first establishment, there were 9 different interruptions with a total power failure time of 26,166 s. This makes approximately 436 min, in other words 7 h and 16 min. When the direct manufacturing loss is calculated on $4,000/h, the total is $29,000. (for 6 months). The efficiency loss in complete re-activation of the establishment after the interruptions should also be calculated. When we consider an approximate manufac-turing loss of half an hour for each interruption, it will make $2000/ interruption. However, in calculation of the number of the interuptions, it will be appropriate that we count the interruptions occurring on the same day as a single one. Thus, if we consider that there are 7 interruptions, the material loss for 6 months will be $14000. The uncontrolled interruption can also lead to breakdowns. The material loss from such breakdowns is calculated to be $3000 for 6 months.

The number of the interruptions in the second establishment was slightly higher than that of the first one. The number of the interruptions in this establishment after the measurement for 6 months was 10. The total duration of such interruptions was 21710 s. In other words, it was around 360 min, corresponding to 6 h. In this establishment, the manufacturing loss per hour was $5000. Given the situation, the material loss due to the duration of the interruption for 6 months was calculated to be $30000. The material loss stemming from the reduced efficiency in re-activation of the establishment can be calculated from manufacturing loss per hour for each interruption in a similar way as the first establishment. This value will be $2500/interruption for this establishment. If we take 9 interruptions in total since the interruptions were successive on 1st August, the calculated loss will be $22500. In this establishment, the uncontrolled interruptions due to the failures cost $2500 for six months. The loss was calculated to be $55000 in total for six months. This means an annual loss of $110,000. The calculations show that it is compulsory that the mea-sures are taken against the failures in both establishments. Voltage flicker Voltage flicker measurements were performed at both

establishments. The results revealed that there were intense poor power quality occurrences in the first esta-blishment. Despite there were voltage flicker recordings in the second establishment, there were no voltage flicker and thus periods exceeding the limits values during the 2 h periods (according to EN50160). The material loss from the voltage flicker suffered by the establishments could not been clearly determined. Since there were several poor voltage quality occurrences in the establishments, it was difficult to make a classification to determine such damage.

Identification would be easier in an establishment where the only poor electrical quality occurrence relates to voltage flicker. In some establishments voltage flicker may directly result in defective products. The voltage flicker values in the first establishment did not cause any directly defected products. Figures 11 and 12 show the graphics of voltage flicker indices from both establishments. As seen from the figures, there were moments when the limits were exceeded.

While the limit value for the voltage flicker parameter Plt was 0.8, the limit value for Pst was 1. In Figures 10 and 11 the values of 0.8 and 1 can be followed by the dashed lines. In the second establishment, both the number of the exceeded limits and the amount of the limits exceeded was lower. It is seen that in the first establishment the values were exceeded much more both in quantity and rate. Total costs The 6-month costs for poor electrical quality in the first establishment were calculated as being $149000 due to voltage sag, $46000 due to failures and $3000 due to transient occurrences, harmonics and voltage flickers. Thus, the total cost for 6 months was $198000, corres-ponding to an annual cost of $396000. In the second establishment, the annual costs were calculated as being $110000 from voltage sags and $5000 from harmonics, making a total of $115000. The losses of the first establishment were higher due to the poor quality of the network and the shortcomings in their own infrastructure. In the second establishment, except for the voltage interruptions, the costs were relatively lower due to the quality of the network and their own infrastructure. Conclusions As demonstrated by the measurements, although the harmonic currents are above the standards in the textile industry, nearly all of the material loss suffered due to poor energy quality is from sudden voltage variations (particularly short-term voltage variations) and power failures. The loss incurred by the first establishment is significantly high, being around 15% of the annual net profit of the facility where measurements were performed. On the


Figure 10. Monthly voltage flicker graphic of the first establishment.

Figure 11. Monthly voltage flicker graphic of the second establishment.


Figure 12. Monthly voltage flicker graphic of the second establishment

other hand, there was no significant cost in the second establishment, except for the power failures. Despite this, it appears that measures should be taken against the loss from failures. Textile industry is a comprehensive sector with various processes carried out in different machinery parks. Therefore, the losses from poor power quality will vary depending on the establishment. While founding a new establishment the investment feasibility parameters should include the high quality electrical power costs along with land costs, qualified staff costs, etc. The fact that the loss incurred by the first establishment is nearly four times greater than that of the second one which utilizes higher quality energy.

The textile sector inour country is troubled with serious global competition and sustaining its existence. There-fore, the government produces incentives and support packages to reinforce this sector which provides employment for around 3 million people, and the main condition for benefiting from such packages is to transfer the establishments to the Eastern or Southeastern Region. The industrialists to transfer their establishments to those regions should take the power quality into consideration, if it is to have a serious effect on the processes and the losses.

The next study will focus on the efficiency losses due to the poor electrical energy quality and proposed solutions to reduce or eliminate the osts. Depending on the costs incurred, the feasibility of the system investments will be

considered within the scope of the proposed solutions. ACKNOWLEDGEMENTS The authors would like to extend their thanks to Schneider Electric Turkey and Mr. Gürkan Erdeniz and also thank Mr. Mustafa Beker for his contributions. REFERENCES Alkan A, Yilmaz AS (2006). “Frequency Domain Analysis of Power

Systems Transients using Welch and Yule Walker Ar Methods”, Energy Conversion Manage. , 7: 2129-2135.

Davis EJ, Emanuel AE, Pileggi DJ (2000). “Evaluation of Single Point Measurement Method for Harmonic Pollution Cost Allocation”, IEEE Trans. Power Delivery, 15 (1): 14-18.

De Abreu JPG, Emanuel AE (2002). Induction Motor Thermal Aging Caused by Voltage Distortion and Imbalance: Loss of Useful Life and Its Estimated Cost, IEEE Trans Indus. Appl., 38 (1): 12-20.

De Nooij M, Koopmans C, Bijvoet C (2007). The Value of Supply Se-curity The Costs of Power Interruptions: Economic Input for Damage Reduction and Investment in Networks, Energy Econs., 29: 277-295.

EN Std 50160 (1999). Voltage Characteristics of Electricity Supplied by Public Distribution Systems EN Standards.

Grupta CP, Milanovic JV, Aung MT (2004). The Influence of Process Equipment Composition on Financial Losses due to Voltage Sags, 11th International Conference on Harmonics and Quality of Power.

IEEE Std 1159 (1995). Recommended Practice for Monitoring Electric Power Quality”, IEEE Standards, IEEE Press.

IEEE Std 1159.3 (2003). “Recommended Practice for the Transfer of Power Quality Data”, IEEE Standards, IEEE Press.

Koçyi�it F, Yanıko�lu E, Yilmaz AS, Bayrak M (2008). Energy Quality

Measurements in a Textile Industry and the Cost of Poor Energy Quality for he Facility (in Turkish), Eleco Congress, Bursa, Turkey.

Sözen A, Bayrak, M, Ye�in EM (2007). Cost of Poor Energy Quality in a Paper Industry, (in Turkish), Energy Efficiency and Quality Symposium, Kocaeli, Turkey.

Kocyigit et al. 1099 Sullivan MJ, Vardell T, Suddeth BN, Vojdani A (1996). Interruption

Costs, Customer Satisfaction and Expectations for Service Reliability”, IEEE Transactions on Power Systems, 11 (2): 989-995.