28 High Voltage Engineering:Fundamentals 5 Figure 2.10 Diagrammatic cross-section of the Felici generator. (1)Cylindrical stator.(2)Insulating rotor.(3)Ionizer.(4)Contact metallic segments.(5)Auxiliary generator.(6)Load.(7)Stationary insulating core (-V=30kV;+V=200kV) is shown in Fig.2.11.The distinctive features include a cylindrical pressure vessel enclosing the generator,the rotor of which is driven at 3000 rpm by means of an induction motor.Ions from an exciting source are sprayed onto the rotor at the charging poles and are transported to the output poles with a consequent rise of potential.The transfer of charge takes place by means of thin blades placed a short distance from the rotor,and in the absence of any rubbing contact the efficiency of the machine is about 90 per cent.The char- acteristics of the 600-kV generator are such that the fluctuation in the voltage is less than 10-4 per cent and the voltage drop at full load current of 4mA is only 500 V.For a 5 per cent variation in the main voltage,the generator voltage remains within 10-5 per cent. The main applications of these 'rotating barrel'generators are in physics as well as in different areas of industrial applications,but rarely in h.v.testing. The maximum voltages are limited to less than 750kV. Finally,another type of electrostatic generator is the vacuum-insulated 'varying capacitance machine',first discussed in detail by Trump(18)and recently again investigated by Philp.(This machine provides a high voltage in the range up to about 1 MV and/or high power in the range of megawatts. The high efficiency,however,could only be reached by high field gradients within the generator,which up to now can only be obtained theoretically by assuming the possible high E values in vacuum.It is,however,doubtful whether the stresses necessary can be reached within the large electrode areas
28 High Voltage Engineering: Fundamentals 4 +V −V 1 2 3 7 5 6 + + + + + + − − − − − − Figure 2.10 Diagrammatic cross-section of the Felici generator. (1) Cylindrical stator. (2) Insulating rotor. (3) Ionizer. (4) Contact metallic segments. (5) Auxiliary generator. (6) Load. (7) Stationary insulating core (V D 30 kV; CV D 200 kV) is shown in Fig. 2.11. The distinctive features include a cylindrical pressure vessel enclosing the generator, the rotor of which is driven at 3000 rpm by means of an induction motor. Ions from an exciting source are sprayed onto the rotor at the charging poles and are transported to the output poles with a consequent rise of potential. The transfer of charge takes place by means of thin blades placed a short distance from the rotor, and in the absence of any rubbing contact the efficiency of the machine is about 90 per cent. The characteristics of the 600-kV generator are such that the fluctuation in the voltage is less than 104 per cent and the voltage drop at full load current of 4 mA is only 500 V. For a 5 per cent variation in the main voltage, the generator voltage remains within 105 per cent. The main applications of these ‘rotating barrel’ generators are in physics as well as in different areas of industrial applications, but rarely in h.v. testing. The maximum voltages are limited to less than 750 kV. Finally, another type of electrostatic generator is the vacuum-insulated ‘varying capacitance machine’, first discussed in detail by Trump18 and recently again investigated by Philp.19 This machine provides a high voltage in the range up to about 1 MV and/or high power in the range of megawatts. The high efficiency, however, could only be reached by high field gradients within the generator, which up to now can only be obtained theoretically by assuming the possible high E values in vacuum. It is, however, doubtful whether the stresses necessary can be reached within the large electrode areas
Generation of high voltages 29 Output cable Voltage divider resistor chain Series resistor chain Collecting electrodes -Spraying Inductor electrode Inductor Stator Frame- Glass Rotor cylinder Figure 2.11 Sames electrostatic generator present in such generators,and therefore only a reference to this type of generator might be useful. 2.2 Alternating voltages As electric power transmission with high a.c.voltages predominates in our transmission and distribution systems,the most common form of testing h.v. apparatus is related to high a.c.voltages.It is obvious then that most research work in electrical insulation systems has to be carried out with this type of voltage. In every laboratory HVAC supplies are therefore in common use.As far as the voltage levels are concerned,these may range from about 10kV r.m.s.only up to more than 1.5 MV r.m.s.today,as the development of
Generation of high voltages 29 Series resistor chain Output cable Voltage divider resistor chain Inductor Frame Glass cylinder Collecting electrodes Spraying electrode Inductor Rotor Stator Figure 2.11 Sames electrostatic generator present in such generators, and therefore only a reference to this type of generator might be useful. 2.2 Alternating voltages As electric power transmission with high a.c. voltages predominates in our transmission and distribution systems, the most common form of testing h.v. apparatus is related to high a.c. voltages. It is obvious then that most research work in electrical insulation systems has to be carried out with this type of voltage. In every laboratory HVAC supplies are therefore in common use. As far as the voltage levels are concerned, these may range from about 10 kV r.m.s. only up to more than 1.5 MV r.m.s. today, as the development of
30 High Voltage Engineering:Fundamentals transmission voltages up to about 1200kV has proceeded for many years. For routine testing,the voltage levels for power-frequency testing are always related to the highest r.m.s.phase-to-phase voltage Vm of power transmission systems.This 'rated power-frequency short duration withstand voltage'V,is different for different apparatus used within the transmission systems and also dependent upon the type of insulation coordination applied (see Chapter 8). For Vm <300kV,the ratio V/V is up to about 1.9 and may decrease with higher values of Vm.If,nevertheless,higher nominal voltages for the a.c. testing supplies are foreseen,the necessity for the determination of safety factors are most responsible for this fact. In general,all a.c.voltage tests are made at the nominal power frequency of the test objects.Typical exceptions are related to the testing of iron-cored windings,i.e.potential or instrument transformers,or to fundamental studies on insulating materials or systems.For iron-cored windings,the frequency has to be raised to avoid saturation of the core.Depending upon the type of testing equipment used,the methods for the generation of variable-frequency voltages might be expensive. A fundamental design factor for all a.c.testing supplies is an adequate control system for a continuous regulation of the high output voltages.In general,this will be performed by a control of the primary or 1.v.input of the voltage step-up systems.It is not the aim of this chapter to deal with the details of these systems.Some hints related to the different methods will be given in section 2.4. Although power transmission systems are mostly of three-phase type,the testing voltages are usually single-phase voltages to ground.The waveshapes must be nearly pure sinusoidal with both half-cycles closely alike,and according to the recommendations(2.3)the results of a high-voltage test are thought to be unaffected by small deviations from a sinusoid if the ratio of peak-to-r.m.s.values equals v2 within +5 per cent,a requirement which can be assumed to be met if the r.m.s.value of the harmonics does not exceed 5 per cent of the r.m.s.value of the fundamental.The r.m.s.value is for a cycle of T V(r.m.s.)= V2(r)dt. J0 The nominal value of the test voltage,however,is defined by its peak value divided by 2,i.e.Vpeak/2.The reason for this definition can be found in the physics of breakdown phenomena in most of the insulating materials,with the breakdown mainly following the peak voltages or the highest values of field strength. Testing of h.v.apparatus or h.v.insulation always involves an application of high voltages to capacitive loads with low or very low power dissipation
30 High Voltage Engineering: Fundamentals transmission voltages up to about 1200 kV has proceeded for many years. For routine testing, the voltage levels for power-frequency testing are always related to the highest r.m.s. phase-to-phase voltage Vm of power transmission systems. This ‘rated power-frequency short duration withstand voltage’ Vt is different for different apparatus used within the transmission systems and also dependent upon the type of insulation coordination applied (see Chapter 8). For Vm < 300 kV, the ratio Vt/Vm is up to about 1.9 and may decrease with higher values of Vm. If, nevertheless, higher nominal voltages for the a.c. testing supplies are foreseen, the necessity for the determination of safety factors are most responsible for this fact. In general, all a.c. voltage tests are made at the nominal power frequency of the test objects. Typical exceptions are related to the testing of iron-cored windings, i.e. potential or instrument transformers, or to fundamental studies on insulating materials or systems. For iron-cored windings, the frequency has to be raised to avoid saturation of the core. Depending upon the type of testing equipment used, the methods for the generation of variable-frequency voltages might be expensive. A fundamental design factor for all a.c. testing supplies is an adequate control system for a continuous regulation of the high output voltages. In general, this will be performed by a control of the primary or l.v. input of the voltage step-up systems. It is not the aim of this chapter to deal with the details of these systems. Some hints related to the different methods will be given in section 2.4. Although power transmission systems are mostly of three-phase type, the testing voltages are usually single-phase voltages to ground. The waveshapes must be nearly pure sinusoidal with both half-cycles closely alike, and according to the recommendations2,3 the results of a high-voltage test are thought to be unaffected by small deviations from a sinusoid if the ratio of peak-to-r.m.s. values equals p2 within š5 per cent, a requirement which can be assumed to be met if the r.m.s. value of the harmonics does not exceed 5 per cent of the r.m.s. value of the fundamental. The r.m.s. value is for a cycle of T Vr.m.s. D 1 T T 0 V2t dt. The nominal value of the test voltage, however, is defined by its peak value divided by p2, i.e. Vpeak/ p2. The reason for this definition can be found in the physics of breakdown phenomena in most of the insulating materials, with the breakdown mainly following the peak voltages or the highest values of field strength. Testing of h.v. apparatus or h.v. insulation always involves an application of high voltages to capacitive loads with low or very low power dissipation
Generation of high voltages 31 only.In general,power dissipation can be completely neglected if the nominal power output of the supply is determined.If C,is the capacitance of the equip- ment or sample under test,and V the nominal r.m.s.voltage of the h.v.testing supply,the nominal KVA rating P may be calculated from the design formula Pn =kV2@C (2.14) in which the factor k I accounts for additional capacitances within the whole test circuit and some safety factor.Examples for additional capacitances are h.v.electrodes and connections between test object and voltage source,which might have large diameters and dimensions to avoid heavy discharges or even partial discharges,or measurement devices as,e.g.,capacitor voltage dividers or sphere gaps frequently incorporated within the test circuit.This safety factor k might range from only about 2 for very high voltages of >1 MV,and may increase to higher values for lower nominal voltages,as over-dimensioning is economically possible.The capacitance of test equipment C,may change considerably,depending upon the type of equipment.Typical values are: Simple post or suspension insulators some 10pF Bushings,simple and graded 100-1000pF Potential transformers 200-500pF Power transformers <1000kVA ~1000pF >1000kVA ~1000-10000pF H.V.power cables: Oil-paper impregnated ~250-300pF/m Gaseous insulated ~60 pF/m Metal clad substation,SF6 insulated ~1000->10000pF One may calculate the nominal currents I=Pn/Vn from egn (2.14)for different test voltages,different C,values as shown above,and proper safety factors k.From such estimations it may be seen that these currents may range from some 10mA for testing voltages of 100kV only,up to amperes in the megavolt range.Although these currents are not high and the nominal power is moderate,many efforts are necessary to keep the test equipment as small as possible,as the space is limited and expensive within any h.v.laboratory. Frequently the equipment will be used also for field testing.Then the porta- bility and transportation calls for lightweight equipment.Some facilities are possible by the fact that most of the test voltages are only of short dura- tion.The nominal ratings are,therefore,often related to short time periods of 15 min.Due to the relatively large time constants for the thermal tempera- ture rise,no sophisticated cooling systems are in general necessary within the voltage testing supplies. A final introductory remark is related to the necessity that all supplies can withstand sudden voltage breakdowns of the output voltage.The stress to the
Generation of high voltages 31 only. In general, power dissipation can be completely neglected if the nominal power output of the supply is determined. If Ct is the capacitance of the equipment or sample under test, and Vn the nominal r.m.s. voltage of the h.v. testing supply, the nominal KVA rating Pn may be calculated from the design formula Pn D kV2 nωCt 2.14 in which the factor k ½ 1 accounts for additional capacitances within the whole test circuit and some safety factor. Examples for additional capacitances are h.v. electrodes and connections between test object and voltage source, which might have large diameters and dimensions to avoid heavy discharges or even partial discharges, or measurement devices as, e.g., capacitor voltage dividers or sphere gaps frequently incorporated within the test circuit. This safety factor k might range from only about 2 for very high voltages of ½1 MV, and may increase to higher values for lower nominal voltages, as over-dimensioning is economically possible. The capacitance of test equipment Ct may change considerably, depending upon the type of equipment. Typical values are: Simple post or suspension insulators some 10 pF Bushings, simple and graded ¾100–1000 pF Potential transformers ¾200–500 pF Power transformers <1000 kVA ¾1000 pF >1000 kVA ¾1000–10 000 pF H.V. power cables: Oil-paper impregnated ¾250–300 pF/m Gaseous insulated ¾60 pF/m Metal clad substation, SF6 insulated ¾1000–>10 000 pF One may calculate the nominal currents In D Pn/Vn from eqn (2.14) for different test voltages, different Ct values as shown above, and proper safety factors k. From such estimations it may be seen that these currents may range from some 10 mA for testing voltages of 100 kV only, up to amperes in the megavolt range. Although these currents are not high and the nominal power is moderate, many efforts are necessary to keep the test equipment as small as possible, as the space is limited and expensive within any h.v. laboratory. Frequently the equipment will be used also for field testing. Then the portability and transportation calls for lightweight equipment. Some facilities are possible by the fact that most of the test voltages are only of short duration. The nominal ratings are, therefore, often related to short time periods of 15 min. Due to the relatively large time constants for the thermal temperature rise, no sophisticated cooling systems are in general necessary within the voltage testing supplies. A final introductory remark is related to the necessity that all supplies can withstand sudden voltage breakdowns of the output voltage. The stress to the
32 High Voltage Engineering:Fundamentals windings and coils accompanied by the breakdown events is usually not related to the short-circuit currents and thus the magnetic forces within the windings, as those currents are not large either;more frequently it is the stray potential distribution between the windings which will cause insulation failures.One may also provide proper damping resistors between h.v.testing supply and the test equipment to reduce the rate of the sudden voltage drop and to avoid any overvoltages within the test circuit caused by interruptions of the breakdown phenomena.Nominal values of such damping resistors between 10 and 100 kS will usually not influence the test conditions.These resistors,however,are expensive for very high voltages and it should be checked whether the a.c. voltage supply can withstand the stresses without the damping resistors. Most of the above remarks are common to the two main methods for the generation of high a.c.testing voltages:transformers and resonant circuits. 2.2.1 Testing transformers The power frequency single-phase transformer is the most common form of HVAC testing apparatus.Designed for operation at the same frequency as the normal working frequency of the test objects (i.e.,60 or 50Hz),they may also be used for higher frequencies with rated voltage,or for lower frequencies,if the voltages are reduced in accordance to the frequency,to avoid saturation of the core. From the considerations of thermal rating,the kVA output and the funda- mental design of the iron core and windings there is not a very big difference between a testing and a single-phase power transformer.The differences are related mainly to a smaller flux density within the core to avoid unneces- sary high magnetizing currents which would produce higher harmonics in the voltage regulator supplying the transformer,and to a very compact and well- insulated h.v.winding for the rated voltage.Therefore,a single-phase testing unit may be compared with the construction of a potential transformer used for the measurement of voltage and power in power transmission systems. For a better understanding of advanced circuits,the fundamental design of such 'single unit testing transformers'will be illustrated.Figure 2.12(a)shows the well-known circuit diagram.The primary winding '2'is usually rated for low voltages of <1kV,but might often be split up in two or more windings which can be switched in series or parallel(not shown here)to increase the regulation capabilities.The iron core 'I'is fixed at earth potential as well as one terminal of each of the two windings.Simplified cross-sections of two possible constructions for the unit itself are given in Figs 2.12(b)and (c). In both cases the layout arrangement of core and windings is basically the same.Figure 2.12(b),however,shows a grounded metal tank unit,for which an h.v.bushing '6'is necessary to bring the high voltage out of the tank '5'.Instead of a bushing,a coaxial cable could also be used if this improves
32 High Voltage Engineering: Fundamentals windings and coils accompanied by the breakdown events is usually not related to the short-circuit currents and thus the magnetic forces within the windings, as those currents are not large either; more frequently it is the stray potential distribution between the windings which will cause insulation failures. One may also provide proper damping resistors between h.v. testing supply and the test equipment to reduce the rate of the sudden voltage drop and to avoid any overvoltages within the test circuit caused by interruptions of the breakdown phenomena. Nominal values of such damping resistors between 10 and 100 k" will usually not influence the test conditions. These resistors, however, are expensive for very high voltages and it should be checked whether the a.c. voltage supply can withstand the stresses without the damping resistors. Most of the above remarks are common to the two main methods for the generation of high a.c. testing voltages: transformers and resonant circuits. 2.2.1 Testing transformers The power frequency single-phase transformer is the most common form of HVAC testing apparatus. Designed for operation at the same frequency as the normal working frequency of the test objects (i.e., 60 or 50 Hz), they may also be used for higher frequencies with rated voltage, or for lower frequencies, if the voltages are reduced in accordance to the frequency, to avoid saturation of the core. From the considerations of thermal rating, the kVA output and the fundamental design of the iron core and windings there is not a very big difference between a testing and a single-phase power transformer. The differences are related mainly to a smaller flux density within the core to avoid unnecessary high magnetizing currents which would produce higher harmonics in the voltage regulator supplying the transformer, and to a very compact and wellinsulated h.v. winding for the rated voltage. Therefore, a single-phase testing unit may be compared with the construction of a potential transformer used for the measurement of voltage and power in power transmission systems. For a better understanding of advanced circuits, the fundamental design of such ‘single unit testing transformers’ will be illustrated. Figure 2.12(a) shows the well-known circuit diagram. The primary winding ‘2’ is usually rated for low voltages of 1 kV, but might often be split up in two or more windings which can be switched in series or parallel (not shown here) to increase the regulation capabilities. The iron core ‘l’ is fixed at earth potential as well as one terminal of each of the two windings. Simplified cross-sections of two possible constructions for the unit itself are given in Figs 2.12(b) and (c). In both cases the layout arrangement of core and windings is basically the same. Figure 2.12(b), however, shows a grounded metal tank unit, for which an h.v. bushing ‘6’ is necessary to bring the high voltage out of the tank ‘5’. Instead of a bushing, a coaxial cable could also be used if this improves