6 Design Considerations of Secondary Systems Egyptian Proverb: The wor ‘To be in bed and sleep no ‘To want for one who come not, ‘To try to please and please not things: cis Scott Fitzgerald, Notebooks, 1925 6.1 INTRODUCTION A realistic view of the power distribution systems should be based on “gathering” functions rather than on “distributing” since the size and locations of the customer demands are not determined by the distribution engineer but by the customers, Customers install all ypes of energy-consuming devices that can be connected in every conceivable combination and at times of customers’ choice. This concept of distribution starts with the individual customers and loads and proceeds through several gathering slages where each stage includes various groups of increasing numbers of cus- tomers and their loads, Ultimately the generating stations themselves are reached through services, secondaries, distribution transformers, primary feeders, distribution substations, subtransmission and bulk power stations, and transmission lines In designing 2 system, distribution engineers should consider not only the immediate, that is, short-range, factors but also the long-range problems. The designed system should not only solve the problems of economically building and operating the systems to serve the loads of today but also require a long-range projection into the future to determine the most economical distribution system ‘components and practices to serve the higher levels ofthe customers’ demands, which will then exist ‘Therefore, the present design practice should be influenced by the requirements of the future system, Distribution engineers, who have to consider the many factors, variables, and alternative solu- tions of the complex distribution design problems, need a technique that will enable them to select the most economical size combination of distribution transformers, secondary conductors, and ser- vice drops (SDs) ‘The recent developments in high-speed digital computers, through the use of computer pro- ‘grams, have provided (1) the fast and economic consideration of many feasible alternatives and {@) the economic and engineering evaluation of these alternatives as they evolve with different strat ‘egies throughout the study period. The strategies may include, for example, cutting the secondary, changing the transformers, and possibly adding capacitors. ‘Naturally, each designed system should meet a specified performance criterion throughout the study period. The most optimum, that i, most economical, system design that correspondls(o a load-growth projection schedule can be selected. Also, through the periodic use ofthe programs, distribution engi- neers can determine whether strategies adopted continue to be desirable or whether they require some ‘modification as a result of some changes in economic considerations and load-growth projections, ‘To minimize the secondary-circuit lengths, distribution engineers locate the distribution trans- formers close to the load centers and try to have the secondary SDs to the individual customers as short as possible, Since only a small percentage of the total service interruptions are due to failures in the second- ary system, distribution engineers, in their system design decisions of the secondary distribution, an332 Electric Power Distribution Engineering are primarily motivated by the considerations of economy, copper losses (FR) in the transformer and secondary circuit, permissible voltage drops, and voltage flicker ofthe system. Of course, there are some other engineering and economic factors affecting the selection of the distribution trans- former and the secondary configurations, such as permissible transformer loading, balanced phase Toads for the primary system, investment costs (ICs) of the various secondary system components ‘cost of labor, cost of capital, and inflation rates. Distribution transformers represent a significant part ofthe secondary system cost, Therefore, one ‘of the major concerns of distribution engineers is to minimize the investment in distribution trans- formers, In general, the present practice in the power industry isto plan the distribution transformer Toading on the basis thal there should not be excessive spare capacity installed, and transformers should be exchanged, or banked, as the secondary load grows, ‘Usually, atransformer load management (TLM) system is desirable for consistent loading practices and economical expansion plans. Distribution engineers, recognizing the impracticality of obtain- ing complete demand information on all customers, have attempted to combine a limited amount ‘of demand data with the more complete, and readily available, energy consumption data available in the customer account files. A typical demand curve is sealed according to the energy consumed, and the resultant information is used to estimate the peak loading on specific pieces of equipment, , feeders, and substations [3-6] such as distribution transformers, in which case is known as TL However, in general, residential, commercial, and industrial customers are categorized in cus- tomer files by rate classification only; that is, potentially useful and important subclassifications are not distinguished. Therefore, the demand data are generally collected for the purpose of generating \ypical curves only for each rate classification, 6.2 SECONDARY VOLTAGE LEVELS Today, the standard (or preferred) voltage levels for the electric power systems are given by the American National Standards Institute's (ANSI) Standard C84.1-1977 entitled Voltage Ratings for Electric Power Systems and Equipment (60 H). Accordingly, the standard voltage level fr single-phase residential loads is 120/240 V. ILis supplied through three-wire single-phase services, from which both 120 V lighting and 240 V single-phase power connections are made to large household appliances such as ranges, clothes dryers, and water healers. For grid- or mesh-type secondary-network systems, used usually in the areas of commercial and residential customers with high-load densities, the voltage level is 208Y/120 V. Itis also supplied through three-wire single-phase services, from which both 120 V lighting and 208 V single-phase power connections are made. For “spot” networks used in downtown areas for high-rise buildings with superhigh-load densities and also for areas of industrial and/or commercial customers, the voltage level s 480¥/27 V. Ivis supplied through four-wire three-phase services, from which both 277 V for fluorescent lighting and other single-phase loads and 480 V three-phase power connections are made, ‘Today, one can also find other voltage levels in use contrary to the ANST standards, for example, 120/240 V four wire three phase, 240 V three wire three phase, 480 V three wire three phase, 240/416 V four wire three phase, or 240/480 V four wire three phase. To increase the service reliability for critical loads, such as hospitals, computer centers, and ‘crucial industrial loads, some backup systems, for example, emergency generators and/or batteries, with automatic switching devices are provided, 6.3. PRESENT DESIGN PRACTICE ‘The part of the electric utility system that is between the primary system and the consumer's property is called the secondary system, Secondary distribution systems include step-down distri- bution transformers, secondary circuits (secondary mains), consumer services (or SDs), and meters, to measure consumer energy consumption.Design Considerations of Secondary Systems 333 Generally, the secondary distribution systems are designed in single phase for areas of residential ‘customers and in three phase for areas of industrial or commercial customers with high-load densi- lies. The types of the secondary distribution systems include the following: 1. The separate-service system for each consumer with separate distribution transformer and secondary connection 2. The radial system with a common secondary main, which is supplied by one distribution transformer and feeding a group of consumers 3. The secondary-bank system with a common secondary main that is supplied by several distribution transformers, which are all fed by the same primary feeder 4. The secondary-network system with a common grid-type main that is supplied by a large number of the distribution transformers, which may be connected to various feeders for their supplies ‘The separate-service system is seldom used and serves the industrial- or rural-lype service areas. Generally speaking, most of the secondary systems for serving residential, rural, and light- ‘commercial areas are radial designed. Figure 6.1 shows the one-line diagram of a radial secondary system. Ithas a low cost and is simple to operate, Dissbution substation 12.47 KV bus L] Feeder circuit breaker Primary feeder main Lateral (Primary fase) Distribution ON transformer 120)240V Secondary circuit (ccondary mai) (Serving #20 hoses) FIGURE 6.1. One-line diagram ofa simple radial secondary system.334 Electric Power Distribution Engineering 6.4 SECONDARY BANKING ‘The “banking” of the distribution transformers, that is, parallel connection, or, in other words, interconnection, of the secondary sides of wo or more distribution transformers, which are sup- plied from the same primary feeder, is sometimes practiced in residential and light-commercial areas where the services are relatively close to each other, and therefore, the required spacing between transformers is litle, However, many ulilities prefer to keep the secondary of each distribution transformer separate from all others. Ina sense, secondary banking isa special form of network configuration on a radial distribution system, The advantages of the banking of the distribution transformers include the following: 1. Improved voltage regulation 2. Reduced voltage dip or light flicker due to motor starting, by providing parallel supply paths for motor-starting currents 3, Improved service continuity or reliability 4, Improved flexibility in accommodating load growth, at low cost, that is, possible increase in the average loading of transformers without corresponding increase in the peak load [Banking the secondaries of the distribution transformers allows us to take advantage of the load. diversity existing among the greater number of consumers, which, in turn, induces a savings in the required transformer kilovolt-amperes. These savings can be as large as 35% according to Lokay [7], depending upon the load types and the number of consumers. Figure 6.2 shows two different methods of banking secondaries. The method illustrated in Figure 6.2a is commonly used and is generally preferred because it permits the use of a lower-rated fuse on the high-voltage side of the transformer, and it prevents the occurrence of cascading the fuses, This method also simplifies the coordination with primary-feeder sectionalizing fuses by having a lower-rated fuse on the high side of the transformer. Furthermore, it provides the most ‘economical system. Figure 6.3 gives two other methods of banking secondaries. The method shown in Figure 6.3 is the oldest one and offers the least protection, whereas the method shown in Figure 6.3b offers Primary main Puimary main Primary fuses Primary fuses Distribution transformers Disibution transformers Baked secondary snne Banked J “an Secandary i Secondary fase a @ Serves fo consumers a Services to consumers FIGURE 6.2. Two different methods of banking secondaties: type 1 and () type 2Design Considerations of Secondary Systems 335 Peony mal Primary main Primary Pricnary fuses fuses Distribution Distribution transformers wwansformers AA Secondary Banke ‘secondary mains Secondary TL fuses ‘reakere a TT Tt (Services consumers o Services to consumers FIGURE 6.3 Two additional methods of banking secondaries (a) type 3 and (b) type 4 the greatest protection. Therefore, the methods illustrated in Figures 6.2a and b and 6.3a have some definite disadvantages, which include the following’ 1. The requirement for careful policing of the secondary system of the banked transformers to detect blown fuses. 2. The difficulty in coordination of secondary fuses. 3. Furthermore, the method illustrated in Figure 6.2b has the additional disadvantage of being difficult to restore service after a number of fuses on adjacent transformers have been blown, Today, due to the aforementioned difficulties, many ulilities prefer the method given in Figure 6.3b, The special distribution transformer known as the completely self:protecting-bank (CSPB) transformer bas, in its unit, a builtin high-voltage protective link, secondary breakers, signal lights for overload warnings, and lightning protection, ‘CSPB transformers are built in both single phase and three phase. They have two identical secondary breakers that trip independently of each other upon excessive current flows. In case of a transformer failure, the primary protective links and the secondary breakers will both open, ‘Therefore, the service interruption will be minimum and restricted only to those consumers who are supplied from the secondary section that is in fault, However, all the methods of secondary banking have an inherent disadvantage: the difficulty in performing TLM to keep up with changing load conditions. The main concern when designing a banked secondary system is the equitable load division among the transformers, Its desirable that ‘ransformers whose secondaries are banked in a straight line be within one size of each other. For other types of banking, transformers may be within (wo sizes of each other to prevent exces- sive overload in case the primary fuse of an adjacent larger transformer should blow. Today, in gen- ‘ral, the banking is applied to the secondaries of single-phase transformers, and all transformers in 1a bank must be supplied from the same phase of the primary feeder. 6.5 SECONDARY NETWORKS Generally speaking, most of the secondary systems are radial designed except for some specific service areas (¢g., downtown areas or business districts, some military installations, hospitals) where the reliability and service-continuity considerations are far more important than the cost and.336 Electric Power Distribution Engineering ‘economic considerations, Therefore the secondary systems may be designed in grid- or mesh-type network configurations in those areas, ‘The low-voltage secondary networks are particularly well justified in the areas of high-load den- sity. They can also be built in underground to avoid overhead (OH) congestion. The OH low-voltage secondary networks are economically preferable over underground low-voltage secondary networks in the areas of medium-load density, However, the underground secondary networks give a very high degree of service reliability. In general, where the load density justifies an underground system, it also justifies a secondary-network system, Figure 644 shows a one-line diagram of a small segment of a secondary network supplied by tree primary feeders. In general, the usually low-voltage (208Y/120 V) grid- or mesh-lype second- ary-network system is supplied through network-type transformers by wo or more primary feeders to increase the service reliability. In general, these are radial-type primary feeders. However, the oop-type primary feeders are also in use to a very limited extent, The primary feeders are inter- laced in a way to prevent the supply to any two adjacent transformer banks from the same feeder. As aresult ofthis arrangement, if one primary feeder is out of service for any reason (single contin- ‘gency), the remaining feeders can feed the load without overloading and without any objectionable voltage drop. The primary-feeder voltage levels are in the range of 4.16-34.5 kV. However, there is ‘tendency toward the use of higher primary vollages. ‘Currently, the 15 kV class is predominating, The secondary network must be designed in such a manner as to provide at least one of the primary feeders as a spare capacity together with its trans- formers, To achieve even load distribution between transformers and minimum vollage drop in the network, the network transformers must be located accordingly throughout the secondary network. As explained previously, the smaller secondary networks are designed based upon single con- lingeney, that is, the outage of one primary feeder. However, larger secondary-network systems must be designed based upon double contingency or second contingency, that is, having two feeder dom hd "reiker Soh. UPS sobstiton ow vtage bat FIGURE 6.4 One-line diagram of the small segment ofa secondary-network systemDesign Considerations of Secondary Systems 37 ‘outages simultaneously. According to Reps [7], the factors affecting the probability of occurrence ‘of double outages are as follows: 1. The total number of primary feeders 2. The total mileage of the primary-feeder outages per year 43. The number of accidental feeder outages per year 4. The scheduled feeder-outage time per year 5. The time duration of a feeder outage Even though theoretically the primary feeders may be supplied from different sources such as distribution substations, bulk power substations, or generating plants, itis generally preferred to hhave the feeders supplied from the same substation to prevent voltage magnitude and phase-angle differences among the feeders, which can cause a decrease in the capacities of the associated trans- formers due to improper load division among them. Also, during lightload periods, the power flow in a reverse direction in some feeders connected to separate sources is an additional concern, 6.5.1 SECONDARY MAINS Seelye [8] suggested that the proper size and arrangement of the secondary mains should provide forthe following 1. The proper division of the normal load among the network transformers 2. The proper division of the fault current among the network transformers 3. Good voltage regulation to all consumers 4. Burning off short circuits or grounds at any point without interrupting service All secondary mains (underground or OF) ate routed along the streets and are three phase four wire wye connected with solidly grounded neutral conductor, In the underground networks, the second- ary mains usually consist of single-conductor cables, which may be either metallic or nonmetallic sheathed. Secondary cables commonly have been rubber insulated, but PE cables are now used to a ‘considerable extent. They are installed in duct lines or duct banks. Manholes atthe street intersee- tions are constructed with enough space to provide for various cable connections and limiters and lo permit any necessary repair activities by workers, ‘The secondary mains in the OH secondary networks usually are open-wire circuits with weath- ‘erproof conductors. The conductor sizes depend upon the network-transformer ratings. For a grid \ype secondary main, the minimum conductor size must he able to carry about 60% of the full-load ‘current to the largest network transformer. This percentage is much less for the underground sec- ‘ondary mains. The most frequently used cable sizes for secondary mains are 40 or 250 kemnil and, toa certain extent, 350 and 500 kemil ‘The selection ofthe sizes ofthe mains is also affected by the consideration of burning faults clea, In case of a phase-to-phase or phase-to-ground short circuit, the secondary network is designed lo burn itself clear without using sectionalizing fuses or other overload protective devices. Here, “burning clear” of a faulted secondary-network cable refers to a burning away of the metal forming the contact between phases of from phase to ground until the low voltage of the secondary network ‘can no longer support the arc. ‘To achieve fast clearing, the secondary network must be able to provide for high current values to the fault, The larger the cable, the higher the short-circuit current value needed to achieve the burn- ing clear of the faulted cable. Therefore, conductors of 500 kemil are about the largest conductors used for secondary-network mains, ‘The conductor size is also selected keeping in mind the voltage-drop criterion, so that the voltage drop along the mains under normal load conditions does not exceed a maximum of 3%,338 Electric Power Distribution Engineering 6.5.2. Limrrers ‘Most of the time, the method permitting secondary-network conductors to burn clear, especially in 120/208 V, gives good results without loss of service. However, under some circumstances, particu: Tay at higher voltages, for example, 480 V, this method may not clear the fault due to insufficient fault current, and, asa result, extensive cable damage, manhole fires, and service imterruptions may occur. ‘To have fast clearing of such faults, the so-called limiters are used. The limiter is a high-capacity fuse with a restricted copper section, and iis installed in each phase conductor of the secondary main at each junction point, The limite’s fusing or time—current characteristics are designed to allow the normal network load current to pass without melting but to operate and clear a faulted section of main before the cable insulation is damaged by the heat generated in the cable by the fault current, ‘The fault should be cleared away by the limiters rapidly, before the network protector (NP) fuses blow, Therefore, the time-current characteristics of the selected limiters should be coordinated with the time—current characteristics of the NPs and the insulation-damage characteristics of the cable. ‘The distribution engineer's decision of using limiters should be based upon two considerations: () minimum service interruption and (2) whether the saving in damage to cables pays more than the ‘cost of the limiters, Figure 6,5 shows the time-current characteristics of limiters used in 120/208 V systems and the insulation-damage characteristics of the underground-network cables (paper or rubber insulated) 10000 5 + 1 ny Tnslaton damage 5000 charateriatics —| N balanced creat onan ‘ the pases and necticut 1600 MEM 500 750 NM 50 NEM sto ien. 350 CM seo 300 MC so MM 2 | tsomen =| _ soo ment sk fon tent 30 MeM — aso new | rome 5 [Limiter ling chara 1 109 G00 1609 S900 19900 So000 T0000 Current ia) FIGURE 6.5 Limiter characteristics in terms of time to fuse versus cuttent and insultion-damage char acteristics of the underground-network cables. (Prom Westinghouse Electric Corporation, Electric Utility Engineering Reference Book-Disiribution Systems, Vol. 3, East Pitsburgh, PA, 1965)Design Considerations of Secondary Systems 339 6.5.3 Network PROTECTORS As shown in Figare 6.4, the network transformer is connected to the secondary network through an NP, The NP consists of an air circuit breaker with a closing and tripping mechanism controlled by a network master and phasing relay and backup fuses Al these are enclosed in a metal case, which may be mounted on the transformer or separately ‘mounted, The fuses provide backup protection to disconnect the network transformer from the net- work ifthe NP fails to do so during a fault, The functions of an NP include the following 1. To provide automatic isolation of faults occurring in the network transformer or in the primary feeder. For example, when a fault occurs in one of the high-voltage feeders, it ‘causes the feeder circuit breaker, at the substation, to be open. At the same time, a current flows to the feeder fault point from the secondary network through the network transform- cers normally supplied by the faulted feeder, This reverse power flow triggers the circuit breakers of the NPs connected to the faulty Feeder to open. Therefore, the fault becomes isolated without any service interruption to any of the consumers connected to the network, To provide automatic closure under the predetermined conditions, that is, when the pri- ‘mary-feeder voltage magnitude and the phase relation with respect to the network voltage are correct. For example, the transformer voltage should be slightly higher (about 2 V) than the secondary-network voltage in order to achieve power flow from the network trans- former to the secondary-network system and not the reverse. Also, the low-side transfer voltage should be in phase with, or leading, the network voltage. 3. To provide its reverse power relay to be adequately sensitive to trip the circuit breaker with, ‘currents as small as the exciting current of the transformer. For example, this is important for the protection against line-to-line faulls occurring in unigrounded three-wire primary feeders feeding network transformers with delta connections 4. To provide protection against the reverse power flow in some feeders connected to separate sources. For example, when a network is fed from two different substations, under certain con- ditions, the power may flow from one substation to the other through the secondary network and network transformers. Therefore, the NPs should be able to detect this reverse power flow and t open, Here, the best protection isnot to employ more than one substation asthe souree, As previously explained, each network contains backup fuses, one per phase. These fuses provide backup protection for the network transformer if the NP breakers fail to operate. Figure 6 6 illustrates an ideal coordination of secondary-network protective apparatus, The coor- dination is achieved by proper selection of time delays for the suecessive protective devices placed in series. Table 6.1 indicates the required action or operation of each protective equipment under different fault conditions associated with the secondary-network system. For example, in case of a fault in a given secondary main, only the associated limiters should isolate the fault, whereas in ‘case of a transformer internal fault, both the NP breaker and the substation breaker should trip. Figures 6.4 and 6.7 show three-position switches electrically located at the high-voltage side of the network transformers. They are physically mounted on one end of the network transformer, 6.5.4 Hich-Vourace Swire AAs shown in Figure 67, position 2 is for normal operation, position 3 is for disconnecting the network transformer, and position 1 is for grounding the primary circuit. In any case, the switch is manually operated and is not designed to interrupt current. The fist slep is to open the primary- feeder circuit breaker atthe substation before opening the switch and taking the network unit out of service. Afier taking the unit out, the feeder circuit breaker may be closed to reestablish service to the rest of the network.340 Electric Power Distribution Engineering 10,000 sooof sr Staton beaker 1000 Primary feeder 500 200 ¢ om ioc Ps * J ewok protector % ah c Soyer Sek 10 ose 2 FE ESSE SS SES SHELLS SPs Limiter Current (A) FIGURE 6.6 An ideal coordination of secondary-network overcurrent protection devices. (From Westinghouse Electric Corporation, Elecrric Usilty Engineering Reference Book-Distibution Systems, Val. 3, Bast Pittsburgh, PA, 1965) TABLE 6.1 Required Operation of the Protective Apparatus Fault Type Limiter NP Fuse NP Breaker Substation Circuit Breaker Low-voltage bus Yes Yes No No Transformer itera feult No No Tips ‘Tens Primary feeder No No Dips ‘pe However, the switch cannot be operated, due to an electric imterlock system, unless the network ‘transformer is first de-energized, The grounding position provides safety for the workers during any work on the de-energized primary feeders. To facilitate the disconnection of the transformer from an energized feeder, sometimes a special disconnecting switch that has an interlock with the associated NP is used, as shown in Figure 6.7 Therefore, the switch cannot be opened unless the load is first removed by the NP from the network transformer. 6.5.5 NETWORK TRANSFORMERS In the OH secondary networks, the transformers can be mounted on poles or platforms, depend- ing on their sizes. For example, the small ones (75 or 150 kVA) can be mounted on poles, whereas larger transformers (800 kVA) are mounted on platforms, The transformers are either single-phase ‘or three-phase distribution transformers. In the underground secondary networks, the transformers are installed in vaults. The NP is mounted on one side of the transformer and the three-position high-voltage switch on the other side, This type of arrangement is called a network unitDesign Considerations of Secondary Systems at Peenary feeder High voltage switch to Distribution A senstorset Garren transformer Network ptector fuse 1 1 1 soak} | 1 1 1 1 Nework protecior Limits Tolarge Tend FIGURE 6.7. High-voltage swit Aypical network transformer is three phase, with a low-voltage rating of 216Y/125 V, and can be as large as 1000 KVA. Table 6.2 gives standard ratings for three-phase transformers, which are used as secondary-network transformers. Because of the savings in vault space and in installation costs, network transformers are now built as three-phase units. In general, the network transformers are submersible and oil or askatel cooled. However, because ‘of envizonmental concerns, askarel is not used as an insulating medium in new installations any- ‘more, Depending upon the locale of the installation, the network wansformers can also be ventilated dey type or sealed dry type and submersible. 6.5.6 TRANSTORMER APPLICATION FACTOR Reps [7] defines the application factor as “the ratio of installed network transformer to load.” ‘Therefore, by the same token, Sr Application factor = oD ys. where ‘S75 the total capacity of network ansornes S45, the total load of secondary network342 Electric Power Distribution Engineering TABLE 6.2 ‘Standard Ratings for Three-Phase Secondary: High Voltage Network Transformers Transformer Standard kVA Preferred Ratings for Nominal Taps Low-Voltage system eating of Voltage Rang —IL(AY).—Above Below aeynasV 2aomgoY — 4160 60 None None 00,0, 750 ‘4L60¥2400" None None 4330 None None B00 ove None 4800 000 None '997547501460514500 300,500, 750 no m0" 15 None. ‘onnressniss5a'480 300, $00, 7150 1500 None ss13m 261692916152 nr 300 95 None 11,213/10926/10639110,%352 300,500,750, 1000 19,000 2.000" 85 None 1-7oayt. 400100110809 300, $00,780, 1000 12.500 None 1299/11 875/41 565011250 TnownI2A70Y T,OGOYIISOP 9S None 2.675/12,380/12025/11700 300, $00,780, 1000 15,200 200 85 None 12 g7W/12S40/12,210111.880 300, $00,750, 1000 ‘T620/13,200Y —13,200¥/7620" None 1.grari,seoy12,210017,880 1730 Nove 13.406/130692,719712.375, 7s0¥r940" None 130w/13n69/12,7192.375 440 Heo 95 None 14 040/13680/13,82012,960 $00, $00,750, 1000 23,000 2900 150 2410023500 22.300/2,700 500,750, 1000 2400 25,20028,600 23,40022;800 Source: Westinghouse Hlecui Corporation, Electric Uliy Engineering Reference Book Distribution Stems, ol 8, ast Piaburg, PA, 1965, With ermision, ‘Nove: Allwindins ce dela connected ules otherwise indicated, Prefered ratings that should be wed when establishing new networks High-sotage and low-voltage neural at internally connected by a removable ik ‘The application factor is based upon single contingency, that is, the loss of one of the primary feed- cers, According to Reps (71, the application factor is a function of the following: 1. The number of primary feeders used 2. The ratio of ZZ, where Z,is the impedance of each section of secondary main and Z, is the impedance of the secondary-network transformer 43. The extent of nonuniformity in load distribution among the network transformers under the single contingency Figure 6.8 gives the plots of the transformer application factor versus the ratio of Zy/Z, for different ‘numbers of feeders, For a given number of feeders and a given Zy/Z;tatio, the required capacity of network transformers to supply a given amount of load can be found by using Figure 6.8. 6.6 SPOT NETWORKS A spot network is a special type of network that may have two or more network units feeding a ‘common bus from which services are tapped. The transformer capacity utilization is better in the spot networks than in the distributed networks due to equal load division among the transformers regardless of a single-contingency condition,Design Considerations of Secondary Systems 343 2 am |_| LP [iwe cde a 20 Bas E The feeders Eis] iy 3 [| fie eden bul ob pt [Tren ede 2 u "o 1 z 3 * 5 Zulia FIGURE 6.8 Network-transformer application factors as a function of Zy/Zy ratio and number of feeders used. (From Westinghouse Hlectric Corporation, Electric Urility Engineering Reference Book Distribution Sostems, Vol. 3, Bast Pittsburgh, BA, 1965) ‘The impedance of the secondary main, between transformers, is zero in the spot networks. The spot networks are likely to be found in new high-rise commercial buildings. Even though spot networks with light loads can utilize 208Y/120 V as the nominal low voltage, the commonly used ‘nominal low voltage of the spot networks is 480Y/277 V. Figure 6.9 shows a one-line diagram of the primary system for the John Hancock Center. 6.7 ECONOMIC DESIGN OF SECONDARIES In this section, a method for (at least approximately) minimizing the total annual cost (TAC) of ‘owning and operating the secondary portion of a three-wire single-phase distribution system in a residential area is presented, The method can be applied cither to OH or underground residential distribution (URD) construction, Naturally, it is hoped that a design for satisfactory voltage-drop and voltage-dip performance will agree at least reasonably well with the design that yields mini- ‘mom TAC. 6.7.1 PATTERNS AND SOME OF THE VARIABLES Figure 6.10 illustrates the layout and one particular pattern having one span of secondary line (SL) ‘each way from the distribution transformer. The system is assumed to be built in a straight line along an alley or along rear lot lines. The lots are assumed to be of uniform width d so that each, span of SL is of length 2d. If SLs are not used, then there isa distribution transformer on every pole and OH construction, and every transformer supplies four SDs. ‘The primary line, which obviously must be installed along the alley, is not shown in Figure 6.10, ‘The number of spans of SLs each way from a transformer is an important variable, Sometimes no344 Electric Power Distribution Engineering TERY risers 27 Transformers for He es Switch room ss0¥/277 V LENIN applying FIGURE 6.9 One-line diagram of the muliple primary system for the John Hancock Center, From Westinghouse Blectrie Corperation, Electric Unilty Engineering Reference Book Distribution Systems, Vol, 3, East Pittsburgh, PA, 1965) Pole or underground pad-mousted tubers transformer sD 8D > so ssp Alleyor rearlot line sp so sD sD ea By Pedetalor = 4 4 manhole (en pote or underground) FIGURE 6.10. Ilgsiration of «typical patern SL is used in high-load density areas. In lightload density areas, three or more spans of SL each. ‘way from the transformer may be encountered in practice. IL Figure 6.10 represents an OH system, the transformer, with its arrester(s) and fuse cutout(), is pole mounted. The SL. and the SD may be of either open-wire or triplex cable construction. If Figure 6.10 represents a typical URD design, the transformer is grade mounted on a concrete slabDesign Considerations of Secondary Systems 345 and completely enclosed in a grounded metal housing, or else itis submersibly installed in a hole lined with concrete, Transite, or equivalent material, Both SL and SDs are triplexed or twin concen- trie neutral direct-burial cable laid in narrow trenches, which are backiilled after the installation of the cable. The distribution transformers have the parameters defined in the following ‘Spis the transformer capacity, continuously rated kVA, gis the per unit exciting current (based on S,). Pris the transformer core loss at rated voltage and rated frequency, kW. Pre, isthe transformer copper loss at rated KVA load, KW, ‘The SL has the parameters defined in the following: ‘Ag, is the conductor area, kemil is the conductor resistivity, (Q-emiD/L = 20.5 at 65°C for aluminum cable. ‘The SDs have the parameters Asy and p with meanings that correspond to those given for SLs. 6.7.2 FURTHER AssuMPTIONS 1. All secondaries and services are single phase three wire and nominally 120/240 V. 2. Perfectly balanced loading obtains in all three-wire circuits 3. The system is energized 100% ofthe time, that is, 8760 h/year. 4. The annual loss factor is estimated by using Equation 2.40a, that is, Fis=03Ry + 0.7K, (2.40a) 5. The annual peak-load kilovolt section of SL. or transformer, is estimated by using the maximum diversified demand of the particular number of customers located downstream from the circuit element in ques- tion, This point is illustrated later. 6. Current flows are estimated in kilovoltamperes and nominal operating voltage, usually 240. 7. All loads have the same (and constant) power factor. smpere loading in any element of the pattern, that is, SD, 6.7.3 Generat TAC EQuarion ‘The TAC of owning and operating one pattern of the secondary system is a summation of invest- ‘ment (fixed) costs (ICS) and operating (variable) costs (OCS). The costs to be considered are con- tained in the following equation: macs Sep + Yea! LIC Ltn HOC oer +P OCre tL OCma + POC incr 62 ‘The summations are to be taken for the one standard pattern being considered, like Figure 6,10, but modilied appropriately for the number of spans of SL, being considered. Its apparent that the TAC so found may be divided by the number of customers per pattern so that the TAC can be allocated. ‘on a per customer basis.346 Electric Power Distribution Engineering 6.7.4 luusrratinc THE Assemety oF Cost Data ‘The following cost data are sufficient for illustrative purposes but not necessarily of the accuracy required for engineering design in commercial practice, Some of the cost data given may be quite inaccurate because of recent, severe inflation, The data are intended to represent an OH system using three-conductor triplex aluminum cable for both SLs and SDs, The important aspect of the following procedures is the finding of equations for all costs so that analytical methods can be ‘employed to minimize the TAC: LICy is the annual installed cost of the distribution transformer + associated protective ‘equipment = (250+ 7.26x5;)xi Siransformer 63) where ISKVA SS, S1O0KVA Sis the transformer-rated kVA TC, is the annual installed cost of triplex aluminum SL cable = (60+4,50x Ag, )xi $/1000 ft 4) where Ag, is the conductor area, kemil {is the pu fixed charge rate on investment Note that this cost is 1000 ft of cable, that is, 3000 ft of conductor, 3. ICyy is the annual installed cost of wiplex aluminum SD cable = (60+ 4.50 Agy) xi $/1000 ft 65) In this example, Equations 6.4 and 6.5 are alike because the same material, that is, triplex ‘aluminum cable, is assumed to be used for both SL and SD construction, 4. IC is the annual installed cost of pole and hardware on it but excluding transformer and. transformer protective equipment $160xi $/pole 65) In case of URD design, the cost item ICs, would designate the annual installed cost of a secondary pedestal or manhole. 5. OC is the annual operating cost of transformer exciting current = fac Sp IC Xi Stransfoomer 6m) where IC.,, is the (otal installed cost of primary-voltage shunt capacitors = $5.00/kvar gc i8the an average value of transformer exciting curzent based on S;KVA rating = 0.015 puDesign Considerations of Secondary Systems 347 6. OCyye is the annual operating cost of transformer due to core (iron) losses = Cy, x7+8760xEC 4 Pry S/ttansformer 68) where IC,,, isthe average investment cost of power system upstream, that is, toward generator, from distribution transformers = $350/KVA, EC is the incremental cost of electric energy (oll-peak) 0.008/KWh Pr is the annual transformer core loss, kW }004 x S; 15 KVA SS; < 100 kVA 7. OC, cys the annual operating cost of transformer due to copper losses (5. 2) Pye Slransformer 6% V5: = UC, Xi+ 8760%BC,, Fs) where “a isthe incremental cost of electric energy (on-peak) = SOO10/KWh. Sou i8 the annual maximum KVA demand on transformer Prey isthe transformer copper loss, KW at rated KVA load, =0073+0.00905%5; where 15 kVA SS; $100 kVA 6.10) sis the annual loss factor 8. OC. uis the anaual operating cost of copper loss ina unit length of SL = (C45 + 8160XEC, Fa) Pa.os oly where ‘Pa.uis the power loss in a unit of SL at time of annual peak load due to copper losses, KW Pye. is an PR loss, and it must be related to conductor area Ag, with R= pL/As, (ne sto desde carefully wheter soul represent ng of condotr eng of cable, When eablishing OCs, forthe parclar pate being wed. one has to remember that different sections of SLs may have different values of current and, there- fee diferent Pa 9. 0Ciys thal perating cost of opperlosin aunitlength of SD. OCyys handled like Os dseribed in Bain 61, When developing J” OCs mporanto relate Paes propery othe ol length f SDs in he entire aera 6.7.5 IMWUSTRATING THE ESTIMATION OF CiRcUIT LOADING Te singing astumptons (5) and (6) eter dere one mend ting te Ioaing of tac mettle pa Is ipostent to finden exits he cor enn cach SD inex stn oS. one caer arenas aposinnons wl be or the ppertons cote Cyan JO abd OC348 Electric Power Distribution Engineering TABLE 6.3 IMlustrative Load Data No. of Customers Being Ann, Max. Demand Diversified evArCustomer) 1 so 2 a8 ‘ 30 8 27 0 22 20 a 0 20 100 Ls Sources Lawiece, RE et a, AEE Trans, PAS: 79(p. MD, 199, Jue 1960 Fig. To proceed, itis necessary to have data for the annual maximum diversified kilovolt-ampere demand per customer versus the number of customers being diversified. The illustrative data tabu- Iated in Table 6.3 have been taken from Lawrence, Reps, and Patton's paper entitled Distribution ‘System Planning through Optimized Design, L-Distribution Transformers and Secondaries (13, ‘ig. 3]. As explained in that paper, the maximum diversified demand data were developed with the appliance diversity curves and the hourly variation factors, ‘tis apparent that the data could be plotted and the demand per customer for intermediate num- bers of customers could then be read from the curve, Alternately, if a digital computer is pro- ‘grammed to perform the work described here, a linear interpolation might reasonably be used to cestimate the per customer demand for intermediate numbers of customers, Figure 6.11 shows a pattern having two SLs each way from the transformer. The reader can apply the foregoing data and with linear interpolation find the flows shown in Figure 6.11, The nominal voltage used is 240 V, 2o21--42kVA > sp * »\ > \\ ach so 247 = 19.76 BVA, 4x30212kVA, HS OKVA 3A SUA 2088 FIGURE 6.11 Estimated citcut loading for coppet-toss determinationsDesign Considerations of Secondary Systems 349 6.7.6 Devevorep Torat ANNUAL Cost EQUATION Upon expanding all the cost items (1)=(9) in Section 6:74, taking the correct summations for the pattern being used, and introducing the results into Equation 6.2, one finds that F Gedy +t 12) Ase Le Dx Sp + Bx Agy + 5 where the coetlicients A to H are numerical constants, It is important to note that TAC has been reduced to a function of three design variables, that is, TAC= f(Sp-Azo. and As.) 13) However, one has to remember that many parameters, such as the fixed charge rate , transformer core and copper losses, and installed costs of poles and lines, ate contained in coefficients A to H.1¢ should be further noted thatthe variables S;, Asp, and Ag, ate in fact discrete vatiables. They are not ‘continuous variables, For example, if theory indicates that S,= 31 KVA is the optimum transformer size, the designer must choose rather arbitrarily between the standard commercial sizes of 25 and. 37.5 KVA. The same ideas apply to conductor sizes for Ag, and Ag. 6.7.7 MINimizarion oF tHe Torat Awnuat Costs ‘One may commence by using Equation 6.12, taking three partial derivatives, and setting each deriv- ative to zero: BOAO) 4 1 HAO) _g 8 As. a(TAC) <0 6 es ‘The work required by Equation 6.14 is formidable. The roots of a cubic must be found. At this point, ‘one has the minimum TAC if only S;is varied and similarly for only Ag, and Agp variables. There is no assurance that the true, grant minimum of TAC will be achieved if the results of Equations 6.14 through 6.16 are applied simultaneously “Having in fact discrete variables in this problem, one now discards continuous variable methods. ‘The results of Equations 6.14 through 6.16 are used henceforth merely as indicators of the region that contains the minimum TAC achievable with standard commercial equipment sizes. The prob- lem is continued by computing TAC for the standard commercial sizes of equipment nearest to the results of Equations 6.14 through 6.16 and then for one (or more?) standard sizes both larger and smaller than those indicated by Equations 6.14 through 6.16. ‘The results at this point are a reasonable number of computed TAC values, all close to the ideal- ized, continuous variable TAC. Designers can easily scan these final few TAC resulls and select the (Sp As. and Ay) combinations that they think best350 Electric Power Distribution Engineering 6.7.8 OruER CoNsTRAINTS ‘There are additional criteria that must be met in the total design of the distribution system, whether or not minimum TAC is realized, The further criteria involve quality of utility service Minimum TAC designs may be encountered, which will violate one or more of the commonly used criteria: 1. A minimum allowable steady-state voliage at the most remote service entrance may have been set by law, public utility commission order, or company policy. A maximum allowable motorstatting voltage dip atthe most remote service entrance sim ilarly may have been established, 3. Ordinarily, the ampacity of no section of SLs or SDs should be exceeded by the designer. 4. The maximum allowable distribution transformer loading, in per unit of the transformer ‘continuous rating, should not be exceeded by the designer. Example 6.1 This example deals with the costs of a single-phase OH secondary distribution system in a resi- dential area, Figutes 6.12 and 6.13 show the layouts and the service arrangement to be consid- ‘ered. Note that equal lot widths, hence uniform load spacings, are assumed. All SDs are assumed to be 70 t long, The calculations should be done for one block of the residential area, In case of OH secondary distribution system, assume that there are 12 services per trans formers, that is, there are two transformers per block thal are at poles 2 and 5, as shown in Figure 6.12. Subsequent problems of succeeding chapters will deal with the voltage-drop constraints, which are used to seta minimum standard of quality of service. Naturally, itis hoped that a design for satisfactory voltage-drap performance will agree at least reasonably well with the design for minimum TAC Table 6.4 gives load data to be used in this example problem. Use 30 min annual maximum demands for custome class 2 for this problem, Use the following data and assumptions All secondaries and services ate single phase three wire, nominally 120/240 V. Assume perfectly balanced loading in all single-phase three-wie circuits ‘Assume that the system is energized 100% of the time, that i, 8760 hiyear. Assume the annual load factor to be Fig = 0.35. ‘Assume the annual loss factor to be Fig = 0.3.9 +O.7K, /}_-__swon. sp |sp | sp] sp] so] so] sv | sp |so |sp |so | so EN sap i FIGURE 6.12 Residential area lot layout and service arrangement, Poles lirstsare om)Design Considerations of Secondary Systems 351 sso ae Utility easement alt pe nse il 5 —lsel— Treat J_$—raocts or n.s20 ——] 1oblods or3.300 FIGURE 6.13 Residential area lot layout and utility easement arrangement. TABLE 6.4 Load Data for Example 6.1 No.of Customers 30 Min Ann. Max. Demands (WVA/Customer) Being Diversified ‘Class t Class? Clas 1 180 00 2s 2 as 18 18 4 no 60 1s 2 100 “4 2 Sowce: Fink DG. and Beaty, HW, Standard Handbook for Ele Engineer, Vth dn, MeGraw-Hil, New York, 1978, Figure 3 Note: The kilovoltampere demands cited have been doubled ata i an effort to modernize the data is explained in th reference cited thatthe original msxinm demand data were develop from appliance diver sity eases and hourly variation factor352 Electric Power Distribution Engineering 6. Assume hat the annual peaiclond copper loses are properly evalated S/R) by applying the given class 2 loads as 2. One consumer per SD B. Four consumers per section of SL Tele consumers por bandormer Here, Pacey an PR Toss, and must be related to conductor area Ay with pxt T000%AR where ‘Asis the conductor area, keri pis 20.5 (Q-cmilvit at 65°C for aluminum cable Lis the length of conductor wire involved (not cable length) (th designer must carl establish aconect ain between TOC chat he annual OC perbock, and the amount of Sor which Ps vate) 2. Assume nora operating vlage of 240 V when computing curents 8: Assume 0 0% power factor for bas 5. Aste a fed hare eaptaiation ate of O16. Using the given data and assumptions, evelop a numerical TAC equation applicable to one block of these residential areas for the case of 12 services per transformer, that s, two transformers per ‘lock. The equation should contain the variables ofS, gy, and A... Also determine the following: a, The most economical SD size (Asp) and the nearest larger standard] AWG wite size b. The most economical SL size (Ay) and the nearest larger standard AWG wire size © The most economical distribution transformer size (5; and the nearest larger standard trans- former size d, The TAC per block for the theoretically most economical sizes of equipment «. The TAC per block for the nearest larger standard commercial sizes of equipment {The TAC pet block for the nearest larger transformer size and for the second larger sizes of Ago and Ag, «Fixed charges per customer per month for the design using the nearest larger standard com- mercial sizes of equipment hh. The variable (operating) costs per customer per month for the design using the nearest larger standard commercial sizes of equipment Solution From Equation 6.2, the TAC is taC# Dic) + Peat Vico A Vien t Roca. + Yoe.+ Locra + Locae LOC 2 Since there are two transformers per block and 12 services per transformer, from Equation 6.3, the annual installed cost ofthe two disteibution tanslormers and associated protective equipment i IC, = 20050 + 7.26 x 5) x0 (250 + 7.26 x S)) x 0.15 75+2.178 5; Sfblock 7)Design Considerations of Secondary Systems 353 From Equation 6.4, the annual installed cost of the triplex aluminum cable used for 300 it per transformer (since there is 150 f SL at each side of each transforme in the SLs is Cy, = 2160 + 4.50% Ag) «i = W60-+ 4.50% Ag) 0.15 300MMransformer T0008 5.4+0.405 Ay Syblock (6.18) From Equation 6.5, the annual installed cost of triplex aluminum 24 SDs per block (each SD is 70 ft long) is IG = 2160+ 4.50% Ag) 12x70 fv/SD = 2160+ 4.50% Asp) x0,15% 2160-44 50% Asp)x0.15x2XE =15.12+1.134 Aso $Yolock 19) From Equation 6.6, the annual cost of pole anc! harchvare for the six poles per block is in = $160 Xx 6 polesiblock = 160.0156 = $144/block (6.20) From Equation 6.7, the annual OC of transformer exciting curent per block is oc, Ie $1 % ayy 4 0.015) x 5; $5slvar x 0.15 = 0.0225 5, Siblock (2 From Equation 6.8, the annual OC of core (ion) losses ofthe two transformers per black is OC 4 = 2G + 8760 % EC) 0.004 x 5, = 21§350/kVA x 0.15 + 8760 x $0,008/kWh) 0.004 x 5, = 0.985, Sfolock 622 From Equation 6.9, the annual OC of transformer copper losses of the two transformers per block is OCs =U gate 8760%Ce x 2] wPeeeElectric Power Distribution Engineering where 0.3hip + 0.7’ 0.3(0.35) + 0.7(0.35" = 0.1904 1.= 12 customersiransformer x 4.4 kVA/customer 52.8 kVAfransformer Here, the figure of 4.4 KVA/customer is found irom Table 6.4 for 12 class 2 customers. From Equation 6.10, the transformer copper loss in klowalls al rated kilovoll- ampere load is found as. <= 0.073 + 0.009085, Therefore, (OC, = AIS350/RVAIO.15 + 8760%(S0.0VKWH 0.1804) ( VNiransformer ¥ 1» { S28kVAtransormerY 6 675..0.0080555 5 } = BPD 22 sack (6.2 SF (6.23) From Equation 6.11, the annual OC of copper losses in the four SLs is OC yep = 2ICy, XF# 8760 % EC, x Fils where Pisc, is the copper losses in two SLs at time of annual peak load, KWitransformer (ee Figure 6.14) =P xR aL (eva) (ova) 6kVA) (eway eva) Guay > sD Cable length= 150 (Cable length = 150 Wire length =300 Wire length = 300 FIGURE 6.14 Ilustation of the SLsDesign Considerations of Secondary Systems 355 where pel To00% Ay 20.510-cmill/ftx300 ft wite2 To00%As 1231¢ kemilitransformer 123 “Zeewitvanstormer Thus, 60x($0.01kWhi 0.1904) 122 OC .ey = A$350/RVAD«O.IS + 29 Aso from Equation 6.11, the annual OC of copper losses in the 24 SDs is Oyo.) = Wy X14 8760 * ECL. x FidProce (69.179)Payey losses in the 24 secondary drops atthe time of annual peak load, KW where Pan is the copper PRR where px Tov0%A 20.510-cmilf (0 M)x(24 SDyblock) (2 wires/SD) 00% Ass 68.88 (0:-kemilyblock From Table 6.4, the 30 min annual maximum demand! for one SD per one class 2 customer can bye found as 10 KVA. Therefore, (okVA Y 68.88, 1 0.240kV} “Ay “7000 Proce 1258 wiblock a356 Electric Power Distribution Engineering Thus, 119.58 = 69.179 22 stock 25 ‘Substituting Equations 6.17 trough 6.25 into Equation 6.2, the TAC equation can be found as TAC = (75 +2.178%5;)4(5.4 HOA0S Ag) 15.1241 134% Ao) (144 40.0225 %5;) 17.108 | 8,273 +0 onxs) [2872 ah). SF Sy Aiter simplifying, Tac= 2395202180545, 22 28729405 O18 44 134 Ay +8273 (6.26) 7 we 2, By parially ciferentiating Equation 6.26 with respect to Ac, and equating the resultantto 7er0, TAC _y 444 8273 Aw Ae ° from which the most economical service-drop size can be found as Avo (22) isd 5.41 kemil “Therefore, the nearest larger standard AWG wire size can be found from the copper-con- ductor table (See Table A.1) as 1/0, that is, 106,500 cml b. Similaly, the most economical SL size can be found from AIAG _ 4512.00 Ay AL Ay (222) 0.405 =204.99kemil Therefore, the nearest larger AWG wire size is 4/0, that i, 211.6 kemil «The most econamical distribution transformer size can be found from ATA or 3.492 56,340 3S 1805.Design Considerations of Secondary Systems 357 39kVA Therefore, the nearest larger standard transformer size is 50 KVA, dd. By substituting the found values 0f Asp, Ag, and S;into Equation 6.26, the TAC per block for the theoretically most economical sizes of equipment can be found 25 +0.405 (204.99) 3273 FE saiaiblock €. By substituting the found standard values of Avy, As, and 5; into Equation 6.26, the TAC per block for the nearest larger standard commercial sizes of equipment can be found as TAC = 239.52+3.1805x(50)+ 3492 , 28.170. 0)” GON 2018 44 134 0106.5)+ 2273 ae 085) +0.405%(211.6) 844 block {. The second larger sizes of Ay and Ay, are 133.1 kemil and 250 kel, respectively. Therefore, 3273 1134t033.04 BE {& The fixed charges per customer per month for the design using the nearest larger standard commercial sizes of equipment is tac=(Qler+ Dic + Dien + Dem 5 Saad Ran =$19225customeronh hy, The variable (operating) cost per customer per month fr the design using the nearest larger standard commercial sizes of equipment is TAC (Locus Noein NOC a LOCua Loca) hs 28,170 3.492 17,018, 8275 =}0.0225(50) + 0.98(50) +2847 sos G0) 30” 211.6 1065 |2ax12 $1,0084/customex/month Note that the fixed charges are larger than the OCs.338 Electric Power Distribution Engineering 6.8 UNBALANCED LOAD AND VOLTAGES A single-phase three-wire circuit is regarded as unbalanced if the neutral current is not zero, This happens when the loads connected, for example, between line and neutral, are not equal. The result is unsymmetrical current and vollages and a nonzero current in the neutral line. In that case, the necessary calculations ean be done by using the method of symmetrical components, Example 6.2 This example and Examples 6.3 and 6.4 deal with the computation of voltages in unbalanced single-phase three-wire secondary citcuits, as shown in Figure 6.15. Here, bath the rutual-imped- ance methods and the fuxlinkage methods are applicable as alternative methods for computing the voltage drops in the SL. This example deals with the computation of the complex linkages ‘due to the line currents in the conductors a, B, and n, Assume thatthe distribution transformer used for this single-phase three-wire distribution is rated as 7200/120—240 V, 25 kVA, and 60 Hz, and the n and n; turn ratios are 60 and 30, As Figure 6.15 suggests, the two halves of the low- voltage winding ofthe distribution transformer are independently loaded with unequal secondary loads. Therefore, the single-phase three-wire secondaries are unbalanced. The vertical spacing between the secondary wires is as illustrated in Figure 6.16. Assume that the secondary wites ace i, FIGURE 6.15 An unbalanced single-phase three-wire secondary circuit FIGURE 6.16 Vertical spacing between the secondary wiresDesign Considerations of Secondary Systems 359 made of #4/0 seven-strand hard-drawn aluminum conductors and 400 {tof lin length. Use 50°C resistance in finding the line impedances, Furthermore, assume that (1) the load impedances Z, and Z, are independent of voltage, (2) the primary-side voltage is V; = 7272 V and is maintained constant, and (3) the line capaci- lances and transformer exciting current are negligible. Use the given information and develop ‘numerical equations for the phasor expressions of the flux linkages Jy, 4g, and 4, in terms of T, and I, In ather words, find the coefficient mavix, numerically, in the equation ["ss" ei ei (627) wi Solution The phasor expressions ofthe complex fh linkages Z, J, and J, due tothe line currents inthe conductors 2, and.n can be writen as" aero (Txtn 628 Be =2«107{T xin (6.29) 10 (xin (630) Since Teh-E=0 3) the current inthe neutral conductor can be writen as haT-E (6.32) “Thus, substituting Equation 6.32 into Equations 6.26 through 6.30, Baro [Bein Be shin: 33) R= 2010" (xine oT xn (634) De Z 2s fain) WOT 39 The notation Tn ord for “Tog tothe base360 Electric Power Distribution Engineering ‘Therefore, from Equations 6.33 through 6.35, 25107 xIn 22425107 xn z he oe | B|-]2<107 xin + 2x10°7 xin 636 he B, z 24107 xIn 2 4 24107 xn Du De Thus, from Equation 6.36, the coefficient matrix can be found numerically as 2x10 inc ant07 xin! Tos cooficient | |-extorsid 2x10? xin a2s2x107 0 Wo-t 1.386210" 9,6855%107 8.292107 -9.6855x107] ™ Note that the elements in the coefficient matrix can be converted to weber-teslas per foot i they are multiplied by 0.3048 mt Example 6.3 Assume that, in Example 6.2, 7, fi, and V7, ate specified but not the load impedances 2, and 2. Develop symbalic equations that will give solutions for the load voltages V, V,, and Vy in terms ‘ofthe voltage V, the impedances, and the fax linkages. Solution Since the transformation ratio ofthe distribution transformer is Ala Bb _ 7200 T20V oo the primary-side curent can be written as (637 Here, v2, 638)Design Considerations of Secondary Systems 361 Substituting Equation 6.37 into Equation 6.38, 639) Aso, (6.40) (ean (oa) Substituting Equation 6.41 into Equation 6.42, HZ ° 6.33) ‘so, by writing a second loop equation, ZAR +Rh +ja% Ts +R +h)~j0, =0 (644) Substituting Equation 6.47 into Equation 6.45, (6.5) 4% 40) [i+ 22.1 R4Z,|h + jot -Z) 647)362 Electric Power Distribution Engineering Example 6.4 Assuming that in Example 6.3 the given voltages are 27220 2ozorv and the load impedances are 0.80 + j0.60.2 0.80 + 0.60.0 7 =14.5152+ j19.906560 7, =0,008064 + jo.00276480 determine the following: a The secondary currents T, and, by The secondary neutral current «The secondary voltages V, and Vy . The secondary voltage Vas Solution From Equation 6.43, (6.430) (648) Similarly, from Equation 6.45, (6.49)Design Considerations of Secondary Systems 363 Substituting the given values into Equation 6.48, 0.008064 + 0.027648 +0.8+/0.6 7272 _7 [14.5152 60 60" 21400¥0.486) _ | (400N0.486) 14.5152 _, 19.90656 5280 60 60" 3280 +/37710,3048):400)%1 8.2997, «9.686%) 121.2 = [10.8857 + 0.6846) +7(0.03279+ 0.03899) (6.50) ‘Aso, substituting the given values into Equation 6.49, 7272 _ [145152 , 19.90656 _(400V0.486) ar So ee 1.20656 9 gono64— 0.027648. or 60 J377(0.3048)(400)x10" x (1.3867, + 9.686%, + 8.2997, +9.686h) 121.2 =7,(-0.03278- j0.03899) +7,-0.88574+ 0.50267) (650) ‘Therefore, from Equations 6.50 and 6.51 121.2] [0.8857 + j0.6846 0.03279 + 0.03899 - (652) 121.2] |+0.03279 ~j0.03899 0.88574 + 0.50267 By solving Equation 6.52 653) | 89.8347 — 62.393 107.387 ~j62.5885 a From Equation 6.53, the secondary cutrents are 89.8347 — [62.393 109.376 2-34.78°A and = ~107:387 ~ j62.5885 = 124.295 2210.24°A364 Electric Power Distribution Engineering b. Therefore, the secondary neutral current is T Tt 7.3523+/124.9815 A «. The secondary voltages are W=BxZ, =(109.3762-34.78°)0 236.879) = 109.37622.09°V and «Ze 124.295.210.240 2-36.87") 24.2952-6.6°V d. Therefore, the secondary voltage Vs is +s = 109.37622.09" 4124.2952-6.62" =232.9977-2.55°V Example 6.5 Figure 6.17 shows an ac secondary network, which has been adapted from Ref. (7]. The loads shown in Figure 6.17 are in taree-phase kilowatts and kilovars, with a lagging power factor of 0.85, “The nominal voltage is 208 V. All distribution transformers are rated 500 kVA three phase, with 44160 V delta high voltage and 125/216 V wye grounded low voltage. They have leakage imped- ance Z, of 0.0086 + 0.0492 pu based on transformer ratings. ‘All secondary underground mains have copper 3-#4/0 per phase and 3-#3/0 neutral cables in rronmagnetic conduits. The positive-sequence impedance Zy of 500 it of main is 0.181 + j0.115, puan a 1000 kVA base, All primary-feeder circuits are 1.25 min long, Three single-conductor 500 kcmil 5 KV shielded ‘copper PE-insulated underground cables are used at 90° conductor temperature. Their imped- ances within the small area of the network are neglected. The positive-sequence impedance Z; of the feeder cable is 0.01 + j0.017 pu on a 1000 kVA base for 1.25 min long feeders. The ‘approximate ampacities are 473 A for one circuit per duct hank and 402 A for four equally loaded Circuits per duct bank, The bases used are (1) three-phase power base of 1000 kVA; (2} for secondaries, 125/216 V, 2666.7 A, and! 0.04687 O; and (3) for primaries, 2400/4160 V, 138.9 A, and 1728 Q. The standard 125/216 V network-capacitor sizes used are 40, 80, ancl 120 kvar. In tis study, these capacitors are not switched. Ordinarily, itis desized that distribution circuits not get into leading power factor operation during of-peak load periods. Therefore, the total magnetizing vars {generated by unswitched shunt capacitors should not exceed the total magnetizing vars taken by the off-peak load. In this example, the total reactive load is 3150 kvar at peak load, and itis assumed that of-peak load is one-third of peak load, or 1050 kvar. Therefore, a total capacitor size ‘f 960 kvar has been used, It has been distributed arbitaly throughout the network in standarel sizes but withthe larger capacitor banks generally being located at the larger-load buses and at the ends of radial stubs from the network365 Design Considerations of Secondary Systems (S961 "Va'BE eae TA, nd.07201= A og AEST HFT ‘suatsks wounguasi _yoog 2oussofoy Busaoursug Sinn uno. meats a8 sean cit ast I6e TiNy seo ty BM 83208 sents Tver 088 MEL wove sayz esq ocx 7 aq moxTpaxlePY) £9 34ND over ais sean 09 feanacr AMS366 Electric Power Distribution Engineering TABLE 6.5 Bus Voltage Value (pu) Buses Case Case? Cases Cases 0551 0867 08st osis, ossk 087s 08550860 0978 0885 0965 xr 09590976 09510864 ost 0st 0962 087s 958 097s 0968 0828 0560 0877 0865 0525, oss 09st 098 0880 oso 087 09st 0498 Gm eensaee Using the given data, four separate load flow solutions have been obtained for the following ‘operating conditions in the example secondary network: Case 1: Normal switching: Normal foads and all shunt capacitors ae off Case 2: Normal switching: Normal loads and all shunt capacitors are on. Case 3: First-contingency outage: Primary feeder is out. Normal loads and all shunt capaci tors are on, Case 4: Second-contingency outage: Primary feeders 1 and 4 are out. Normal loads andl all shunt capacitors are on, Note that this second-contingency outage is very severe, causing the largest load (at bus 5) to lose two-thirds ofits transformer capacity ‘To make a voltage study, Table 6.5 has been developed based on the load flow studies for the four cases. The values given in the table are per unit bus voltage values, Here, the buses selected for the study are the ones located at the ends of radials or else the ones that are badly disturbed by the second-contingency outage of case 4, Use the given data and determine the following: I the lowest “favorable” and the lowest “tolerable” voltages ave defined as 114 and 111 V, respectively, what are the pu voltages, based on 125 V, that correspond to the lowest favor- able voltage and the lowest tolerable voltage for nominally 120/208Y systems? b. List the buses given in Table 65, for the first-contingency outage, that have (1) less than favorable voltage and (2) less than tolerable voltage. «List the buses given in Table 65, for the secondl-contingency outage, that have () less than favorable voltage and (2) less than tolerable voltage. . Find 2y/Z, 2242), and using Figure 6.8, find the value ofthe “application factor” for this ‘example network ancl make an approximate judgment about the sufficiency ofthe design of this network. Solution ‘The lowest favorable voltage in per uni is 114 V: MAY oo o12pu Tv and the lowest tolerable voltage in peruDesign Considerations of Secondary Systems 367 b. There are no buses in Table 6.5, forthe first-conlingency outage, that have (1) less than favorable voltage or (2) less than tolerable voltage «. For the secondl-contingency outage, the buses in Table 6.5 that have (1) less than favorable voltage are B, C, J, K, R, and S and (2) les than tolerable voltage are, C, J, and K. d, The given transformer impedance of 0.0086 + /0.0492 pu is based on 500 KVA. Therefore, itconresponds to 2, = 0.0172 + /0.0984 pug that is based on 1000 kVA. Therefore, the ratios are Zu 01ST /O.15 Z, 00172= j0.0984 0735 “Thus, irom Figure 6.8, the corresponding average transformer application factor for four feeders can be found as 1.6, To verify this value forthe given design, the actual application factor can be recalculated as Actual application factor = 12! installed networksranstomer capacity Total load 19 transformers «500 kVAftranstorme —s0565]3158 1.5846 Therefore, the design ofthis network is sufficient. 6.9 SECONDARY SYSTEM COSTS AAs discussed previously, the secondary system consists of the service transformers that convert primary voltage to uilization voltage, the secondary circuits that operate at utilization voltage, and the SDs that feed power direclly to each customer. Many utilities develop cost estimates for this ‘equipment on a per customer basis. The annual costs of operating, maintenance, and taxes for a secondary system are typically between 1/8 and 1/30 of the capital cost Tn general, it costs more to upgrade given equipment to a higher capacity than to build to that ‘capacity in the first place. Upgrading an existing SL entails removing the old conductor and install- ing new, Usually, new hardware is required, and sometimes poles and crossarms must be replaced ‘Therefore, usually, the cost of this conversion greatly exceeds the cost of building to the higher- capacity design in the first place, Because of this, T&D engineers have an incentive to look at long-term needs carefully and (o install extra capacity for future growth, Example 6.6 Ithas been estimated that a 12.47 KV OH, three-phase feeder with 336 kcmil costs $120,000/mile, Ithas been also estimated that to build the feeder with 600 keri conductor instead and a 15 MVA ‘capacity would cost about $150,000/mile, Upgrading the existing 9 MVA capacity line Inter to368 Electric Power Distribution Engineering 15 MVA capacity entails removing the old conductor and installing new. The cost of upgrade is $200,000/mile. Determine the following 2. The cost of building the ® MVA capacity line in dollars per KVA-mile b. The cost of building the 15 MVA capacity line in dollars per KVA-mile cc. The cost ofthe upgrade in dollars per kVA-mile Solution a, The cost of building the 8 MVA capacity line is $120,000 ‘oStywaine = Se Amie Cost, Sateen “STS S3IKVAm 1 of building the 15 MVA capacity is $150,000 3150.000 ss /vAmile OOKVA s Costs pate «The cost ofthe upgrade is $200,000 $200,000 _ 635. s3/kVAmile 5,000-9, 0007kVA Costepgnte {As itcan be seen earlier, when judged against the adeltional capacity (15 ~ 9 MVA\, the upgrade ‘option is very costly, thats, over $33/kVA-mile. PROBLEMS 6.1 Repeat Example 6.1. Assume that there are four services per transformer, that is, one trans- former on each pole so that there are six transformers per block. 6.2 Repeat Example 6.1. Assume that the annual load factor is 0.35 6.3 Repeat Problem 6.1. Assume that the annual load factor is 0.65. 6.4 Consider Problem 6.1 and find the following: The most economical service-drop size (As, and the nearest larger commercial wire size bb. ‘The most economical SL size (Ag,) and the nearest larger standard transformer size . The TAC per block for the nearest larger standaed sizes of equipment 6.5 Repeat Example 6.4, assuming the load impedances are Z,=10+j000 and =15+j000 6.6 Repeat Example 6.4, assuming the load impedances are =104j009 and 3.04000