Background of the Invention

The invention relates generally to a method and apparatus for reducing the size of utility company switch gear and protecting the sensitive components, such as generators and transformers used by power utilities, and in particular, to a current limiting method and apparatus for controlling the current through such generators and transformers during a fault condition.

In the utility industry, to protect current sensitive components such as generators and transformers, the utilities typically use passive current limiters without the ability to trigger upon the occurrence of a fault. Accordingly, these limiters limit the current only to values which are ten to one hundred times higher than the normal operating values for the equipment.

Since the expense and size of switch gear used by the utilities increases rapidly with fault current ratings, the utilities prefer to use the smallest switch gear which is commensurate with the application. The addition of fast operating current limiters, with tight overload specifications, would allow the utilities to split a high current bus without the requirement of using switch gear on each sub-bus which tolerates a high fault current. For these split-bus services, typically, for example, 240kV, 4 kA lines, to operate successfully, the fault currents need to be limited to three or four times the normal load current. Thus, a limiter operating for 30 or 40 milliseconds, on a 240kV line, must be able to sustain a 75-100 MJ energy load per fault.

When a fault occurs, for example a short circuit, the current through the current sensitive components can rise to fault values in less than 1 millisecond. Accordingly, a current limiter to be effective must be able to hold the current to less than several times normal values for at least about 30 milliseconds while the line circuit breaker

/01909

-2- operates. The line circuit breaker, while effective to decouple the fault from the current sensitive components, will not "kick in" for about 30 milliseconds once the fault occurs.

In the past, there have been two forms of current limiters used in utility power systems. These are inductive limiters and current limiting fuses. The current limiting inductors are air core inductors, having an inductance (L), placed in series with a circuit breaker. They limit current by generating a back voltage (V) , due to the changing current (i), as a function of time (t) according to Lenz\'s and Faraday\'s laws:

V = -L t (Eq. 1)

The major advantages of inductive limiters are their simplicity, robustness, and low cost.

Current limiting fuses use a non-linear resistance, R, in series with the line, and the fuse vaporizes when the current exceeds a specified rating. These fuses accordingly are not reusable, and if reusable, would provide a back voltage depending upon the temperature of the limiter in accordance with Equation 2:

V = iR(i,T) (Eq. 2)

The advantages of a current limiting fuse are its low cost and positive action in the case of a fault.

Superconducting variants of these types of current limiters have been built. In addition, transformer type devices have also been built. The superconducting version of the current limiting fuse depends on a large change in resistance from the superconducting to a normally conducting state and the potential that the device can be made

reusable. Some superconducting limiters, operating at voltages lower than that of interest to the power utilities, have also been described in the literature.

Series limiters have the limitation that they must operate at the operating potential. For passive devices, where there is adequate room in the utility switch yard, this is rarely a problem. However, if the device is switchable, then a high voltage transformer-like structure must be built for the trigger signals and the associated electronics. This adds to the complexity and cost of the limiter and active series limiters can also have difficulty in failing "safely." That is, they may not have sufficient heat capacity to prevent melting if a current switch fails to open in the allotted time.

For inductive limiters, the effect of the series inductance during operation needs to be minimal. If the value is small enough so that only a small amount of voltage appears across the inductor during normal operation, there is likely to be only a modest current limiting, for example, after the operating current is greater than ten times its normal value. In addition, there is no capacity to trigger a normal metal inductive limiter.

The current limiting fuse can cause a voltage spike of two or three times the system voltage when the fuse opens. Thus, the insulation required in the system must be adequate to handle the spike. Also, the fuse operates only once before replacement, and replacement can be a time consuming process. The vaporization of melted metal in fuses also limits the upper voltage at which they can be used.

Accordingly, an object of the present invention is an apparatus for limiting current in current sensitive components of a utility distribution process in which the current limiter limits currents to between three and ten times the normal operating current using a triggerable apparatus. Other objects of the invention are an economical, modest size, and reusable method and apparatus for limiting current in power utility switch gear. Yet

another object of the invention is a fast-acting method and apparatus which can be added to existing switch gear equipment in a simple and economical manner for limiting current there through.

Summary of the Invention

The invention relates to a method and apparatus for limiting current transfer in a power transformer. The method features detecting a current fault condition causing dangerously high currents to pass through the transformer and applying sufficient current to the transformer core to bias it into a saturated state for a sufficient time to protect the transformer until operation of a transformer circuit breaker. In another aspect, the method features applying a current pulse to the transformer core for limiting current passing through the transformer using a superconducting coil pulse transformer.

The apparatus of the invention features circuitry for connecting a high current pulse source for saturating the transformer core wherein the source has sufficient energy to effect saturation. The source of pulse energy has a transformer having a superconducting primary winding and further has elements for quenching or rapidly discharging without quenching the superconducting winding in response to a fault detection signal.

Brief Description of the Drawings

Other objects, features and advantages of the invention will be apparent from the following drawings in which:

Figure 1 is a schematic representation of a portion of a utility distribution system; and

Figure 2 is an electrical schematic drawing illustrating a superconducting current limiter connected in accordance with the invention.

Description of a Particular Embodiment

Referring to Figure 1, in a typical utility transmission system 10, the transmission lines 12 nearly always terminate at transformers 14. Transformers have a demonstrated track record of robustness and reliability, and therefore it is advantageous to use the transformers in a manner which provides a device which fails safe, which is mechanically, thermally, and electrically robust, and which does not interrupt or substantially change the operation of the transmission line.

In accordance with the illustrated embodiment of the invention, referring to Figure 2, a transformer 30, has an input side 32, and an output side 34, each connected to a respective primary and secondary winding of the transformer 30. In accordance with a preferred embodiment of the invention, a hybrid pulse power transformer 36 such as that developed by General Dynamics Corporation in San Diego provides a method of pulsing the transformer core, so that the core saturates, thereby limiting the amount of current which can be drawn from the transformer because the transformer is then operating in a different, non-optimum portion of its B-H curve. Biasing the transformer core in this way produces operation very similar to that which can be found in the operation of a magnetic amplifier.

In accordance with the invention, a bias winding 38 is provided through which magnetic core saturation can be achieved upon the detection of a fault condition. A source of bias current is provided to the bias winding from the hybrid pulse power transformer 36 which is triggered in response to a fault condition signal over line 40 to produce a pulse of current upon the detection by a sensor 42 of a fault condition. Sensor 42 measures, for example, the current in the secondary winding of transformer 30.

This method of operation causes the transformer to generate high frequency harmonics, induces current in the transformer case and heats other components of the transformer. If allowed to operate for a long period of

time, this mode of operation could be detrimental to the transformer; however, the large mass of the transformer limits the temperature rise and allows sufficient time for the switch gear to open the circuit (using a circuit breaker, not shown). Since nearly all high power transformers have surge current ratings which provide a margin for continued safe operation, they can typically operate at twice their rated current for periods of minutes to hours, and at three or four times the rated current for tens of seconds to several minutes without suffering damage. This generally results because large transformers are often designed with a substantially more iron (core) than is necessary in order to reduce the hysteresis losses in the iron. This large quantity of iron, however, makes it more difficult to saturate the core and a large current must be applied to winding 38 to induce substantial flux in the core to reduce the short circuit current carrying capacity of the transformer during a fault condition.

In alternate embodiments of the invention, in some core configurations, current can be injected into the transformer core through the transformer neutral. For example, for wye connected transformers, with proper core configuration, the neutral would need to be lifted from ground and the secondary of the hybrid pulse power transformer would be connected in series with the neutral lead of the power transformer. Alternately a separate winding can be added to the core to effect this bias. In a design as described below, the hybrid pulse power transformer will always be fail safe.

The hybrid pulse power transformer used in connection with the invention is illustrated in Figure 2. It is described in greater detail in the manufacturer\'s paper attached as Appendix A. It consists of a superconducting coil 48 with a "quench trigger." The coil can be charged from a power source "charge battery" 49 as illustrated. Once the current i in the superconducting coil has been created, a switch 50 is opened (previously closed) and a

switch 52 is closed (previously opened) as illustrated in Figure 2. So long as the coil is maintained in its superconducting state, no bias current is applied to the transformer core 54. When however a fault condition or state is detected, as indicated by a fault detection signal over line 40 from sensor 42, the superconducting coil 48 is quenched, and the energy stored within it is transferred through the transformer 62 to the core of the transformer 30 which then becomes a saturated, thereby limiting its current carrying capacity. The operation of the pulse transformer as a current source thus provides the ability to trigger the current limiting operating, using existing components, to reuse the pulse transformer, and to provide a fail safe design. Note that the quenching of the superconducting coils is not required. The only requirement is that the current discharge rate from the secondary of the high power pulse transformer is adequate to saturate the transformer core.

The high power pulse transformer contemplated herein can handle up to 100MJ per fault, thus in its circuit configuration, limiting the current through transformer 30 to three to four times its rated capacity. In a preferred embodiment, the quench of coil 48 can take place within one millisecond and the saturation of core 54 will be complete in less than one millisecond. Accordingly, without substantial modification of an existing transmission line, current fault detection and limitation can be achieved at relatively small cost and substantial savings in volume requirements.

Additions, subtractions, and other modifications of the described embodiment will be apparent to those practiced in the field and are within the scope of the following claims.

APPENDIX A

Text re-typed from IEEE Transactions on Magnetics Vol. 25, No. 2, March 1989, pages 1779-1782

"Using a Small Hybrid Pulse Power Transformer Unit as a Component of a High-Current Opening Switch for a Railgun"

E.M.W. Leung, R.E. Bailey, and P.H. Michels

General Dynamics Corporation

Space Systems Division

P.O. Box 85990

San Diego, CA 92138

Abstract The Hybrid Pulse Power Transformer (HPPT) ¹ is a unique concept exploiting the ultra-fast superconducting to normal transition process of a superconductor. When used in the form of a Hybrid Transformer Current Zero Switch (HTCS) , this creates an approach in which the large, high power, high current opening switch in a conventional railgun system can be eliminated. This represents an innovative application of superconductivity to pulsed power conditioning required for the Strategic Defense Initiative (SDI). This paper will explain the working principles of a 100 KJ unit capable of switching up to 500 K-ampere at a frequency of 0.5 Hz and with a system efficiency of greater than 90%. Circuit analysis using a computer code called SPICE PLUS was used to verify the HTCS concept. This concept can be scaled up to applications in the several mega-ampere levels.

Introduction The historical approach to energy compression for SDI railgun applications has been to use a homopolar generator/ storage inductor combination to transfer power between the primary energy source and the load. To accomplish this transfer of energy, two types of switches have been used for

this high-current, opening switch application: direct interrupt and current zero. Direct-interrupt switches experience severe damage and limited lifetime due to unavoidable arcing during the commutation phase. Current- zero switches (switches whose internal currents are reduced to zero by some external means prior to switch opening to minimize contact arcing and erosion), which have been capacitance based, are bulky and inefficient because they are required to withstand high voltages during system operation. Current zero switches have the potential to overcome the limitations of direct interrupting switches if inductive energy storage is used for the required counter current.

A typical inductor-driven pulsed-power delivery system for a railgun system is shown in Figure 1. It consists of a large energy storage inductor (Lj?) and a high voltage and . current direct opening switch. The storage inductor can be charged (charging circuit not shown) with a homopolar generator (HPG) or battery arrays. The opening switch is bulky, heavy, and has to be replaced after a very limited number of firings. The Hybrid Pulse Power Transformer used in the form of a Hybrid Transformer Current-zero Switch concept will remove this bottleneck to continual electromagnetic gun development.

Figure 1. Supplying energy from storage inductor Lg typically requires a high power, high-current opening switch.

The Hybrid Transformer Current-zero Switch (HTCS) Our approach incorporates use of a device we call the Hybrid Pulse Power Transformer (HPPT) . HPPT is used to generate a current pulse in the direction opposite to the current generated by the inductor, momentarily reducing the current through the opening switch to zero. The opening switch opens at this time under essentially no current load and thus, no arcing. The primary advantages of this approach are simplified, smaller opening switches and longer switch life.

Working Principle Of The HPPT The HPPT* is an engineering device being developed to provide mega-ampere level pulsed power without the need of a high power, high voltage breaker switch. Whereas very high voltages are produced during the energy transfer process using a conventional transformer, very low voltages are induced during an HPPT type operation. The HPPT consists of a normal secondary and a superconducting primary coil fabricated using a composite conductor that becomes highly resistive when driven normal. The operational sequence of an HPPT system is as follows:

The primary can be charged and maintained at a level close to the critical current for a relatively long time with minimum energy loss because it is superconducting. When energy transfer is required, a quench induction circuit is engaged to quench the primary rapidly and uniformly. This rapid current collapse in the primary, depending on the turns ratio between the primary and the secondary, can induce up to mega-ampere levels of electrical current in the secondary for delivery to the load.

*Patent applied for.

PRIMARY COIL SEOONDARY COIL(NORMAL)

77 ¹ . , (UNLOADED. .OAD; SECONDARY _ j ^" LOOPOPEN)

(A) CHARGEMODE

^" 7?- , (UNLOADED,

•LOAD- SECONDARY j ^{* J} LOOPOPEN)

(B) QUIESCENT MODE (ENERGY STORAGE)

(C) FIRING MODE (POWER CONDITIONING AND ENERGY TRANSFER)

1. , (SECONDARY LOOP RE- \'LOAE OPENS AFTER LOADING . . Λ CONDITIONCEASES, NOCURRENTN

(D) RECOOUNGMODE PRIMARYNOR SEOONDARY)

Figure 2. An HPPT is a high-efficiency combination energy storage, power conditioning, and energy transfer device.

The HPPT is therefore a combination energy storage, power conditioning and energy transfer device. The time constant for the energy transfer for a given design is determined mainly by the total normal resistance of the primary. Since the resistance and inductance are distributed along the windings, internal inductive and resistive voltage cancellation is locally achieved, resulting in a very low terminal voltage for the transformer. This low voltage also enables the primary and the secondary to be closely coupled, resulting in high energy transfer efficiency.

How An HTCS Works The HTCS consists of a small HPPT and switch. The small HPPT unit is first charged up to a current level close to its critical current (Figure 3). After the power supply for charging up the primary is isolated by opening SW1 and closing SW2 (Figure 4), the HPPT unit is ready for action. After energy storage inductor (L^) is charged up using an external power supply (Figure 5). The system is ready for delivery of power to the load. To fire the railgun, loading the projectile completes the circuit for the secondary current (\'sec) to flow; the quench initiation circuitry is then activated (Figure 6). The key point here is that the anticipated current flow direction for the induced secondary current in the HPPT unit when it is activated will be opposite to the direction of current flow from Lg through the switch SW3. As the current (Isec) ^ ^{n tlιe} hybrid transformer secondary increases, the current passing through the opening switch (SW3) decreases. When the opening switch current (\'zeroSW) ^ ^{s at or} near to zero, switch SW3 can be opened, thereby transferring energy to the load (Figure 7).

SOURCE, THE HOMO¬ POLAR GENERATOR)

Figure 3. The HTCS superconducting primary is charged by closing switch SW1.

Figure 4. Switch SW2 is closed and switch SW1 opened. Power is isolated from HPPT which is now in ready-to-fire mode.

Figure 5. Power supply for energy storage inductor charges the inductor by closing switch SW3.

Figure 8 depicts the conceptual time relationship between the opening switch current and the load current during firing. This represents what we want to get verified by analytical means.

Figure 6, The system is loaded and, simultaneously, the primary of the HPPT is driven normal to produce a secondary current lg, in a direction opposite to that of l _E (IzeroSw = I \'seel " I \'EI)-

Figure 7. As l _sec in the secondary increases, the current through switch SW3 ( I _ze roSw = I \'seel ~ MEI ) decreases. At l _Z eroSW ⁼ 0\' switch SW3 can be opened, thereby releasing stored energy in Lg into load.

Figure 8. Current rise in the load (plasma in back of projectile) begins as soon as the HPPT portion of the HTCS is activated.

Application Of An HTCS To A Railgun System The HTCS concept, integrated into a railgun pulse power system is depicted in Figure 9. This schematic illustrates how a counter current pulse from our HTCS reduces the operating current through the switch to small values. Also noted is the fact that, with this particular series connection, the HTCS actually supplies the initial energy impulse that ignites the plasma at the rear of the projectile. When the switch opens, the operating current from the Homopolar Generator sustains the arc as the projectile moves out of the barrel.

Unique advantages offered by our approach are listed below: a) Our unit is physically small and requires little energy to drive the current through the closed switch to zero.

b ₎ The number of firings before refurbishment of the switch will be much larger than is the case for opening switches currently being used.

ENERGYSTORAGEINDUCTOR

Figure 9. The unique HPPT/HTCS approach enables the switch to open at zero or close-to-zero current.

c) The concept is scalable to larger units giving higher currents. Advanced cooling design can eventually push the repetition rate.

d) Our design will eliminate the need for large, rotating switch designs, and can replace them with, for example, small vacuum contactors or a small slide-type opening switch. This new mechanical switch is estimated to be at least 50% smaller than a conventional one.

e) Compared to a conventional pulse transformer, our HTCS requires minimal electrical insulation due to the self-cancellation of the primary

voltage. A conventional transformer will generate tens of thousands of volts across the primary when the energy is transferred to the secondary. For ours, it would be hundreds instead.

f) The efficiency of energy transfer of our concept (90%) is much higher than for a conventional pulse transformer (50% or lower).

Verification Of The HTCS Working Principle By The Use Of SPICE PLUS Circuit Analysis

Analog Workbench (AWB)^ is a workstation from Analog Design Tools, Inc. that is used for the design and test of analog circuits. It is a collection of programs that operate on a personal computer system that follows the same work flow and procedures as a conventional design environment. Included in the Analog Workbench is SPICE PLUS ⁴ which is a general-purpose circuit simulat; program for nonlinear transient analyses provided by Analw ₃ Design Tools, Inc. It is based directly on SPICE3 which is the latest version of SPICE developed by the University of California at Berkeley. Analog Workbench with SPICE PLUS is used on Apollo workstations available in-house.

Analog Workbench greatly simplifies SPICE analysis and setup by gathering information needed for the input deck for SPICE PLUS as the circuit and test instruments are set up by also providing a four channel function generator, a four channel oscilloscope, and a parametric plotter.

The Symbol Editor in Analog Workbench provides the ability to create symbols for proprietary devices or customize existing symbols. The function generator controls the start and stop times of the simulation and also controls any voltage or current sources. The oscilloscope displays the outputs from selected ammeters or voltmeters placed in the circuit. The parametric plotter is used to plot

families of curves as a result of running analyses with varying component values.

The analysis by SPICE PLUS includes calculating the dc operating point for each circuit node followed by a transient analysis; SPICE PLUS computes the transient output variables as a function of time over a user specified time interval.

The circuit analysis program, SPICE PLUS, used with the Analog Workbench software was used to analyze the circuit shown in Figure 10. Explanation for the symbols is given in the next section and we are basically analyzing the circuit presented in Figures 3 to 7.

Figure 10. Model of the HTCS system used for SPICE PLUS analysis.

Model Set Up For SPICE PLUS Runs

V9 and R44 represent the HPPT primary charging source. U3 represents the resistance of the primary: it is a model of a time varying resistor created by the Symbol Editor in Analog Workbench. The primary resistance U3 is zero until the primary is driven normal, and then it increases to the full normal resistance of the primary.

Transformer X6 represents the HPPT. R23 represents the HPPT secondary resistance. Switch S24 and S25 represent the primary isolation switches (SW1 and SW2 respectively in Figures 3-7).

V8, L30, and R42 represent the homopolar generator source for charging the storage inductor, L28 and its resistance, R39.

R29 and L27 represent the load of the railgun. Switch S22 represents the circuit across the load being unloaded with the switch opened, and then loaded with the switch closed.

Switch S26 represents the opening switch SW3 as shown in Figures 3-7.

Parameters Used In Analysis Primary Inductance Primary Resistance Primary Current Primary Energy Secondary Inductance Secondary Resistance Coupling Coefficient Load Resistance Load Inductance Energy Storage Inductance Energy Storage Resistance Homopolar Generator Source Homopolar Generator Inductance Homopolar Generator Resistance Storage Inductor Current

Analytic Steps Performed a) Initial setup has S24 closed, S25 open, S22 open, and S26 closed. The switches are voltage controlled using the step waveform from the function generator. The simulation time is from zero to 1 seconds.

b) When the homopolar generator has the storage current to its required level (500 KA, at 920 msec), the primary charging source is isolated, the load switch is closed, and the primary is driven normal.

c) At 920.1 msec, when the current through the opening switch is close to zero current, the opening switch (S26) is opened.

1909

-21-

Results

Figure 11 shows the primary charged up to the required current (225 A) and the secondary (load) current at zero before quench. At 920 msec, the primary is isolated from the primary source and the primary is quenched. At 920.1 msec, the opening switch is opened and the stored energy is released into the load. Figure 12 shows an expanded display of the primary current and secondary current from 919.95 to 920.2 msec. Marker 1 is set at 920.1 msec, which shows a primary current value of 2.3 A and a secondary (load) current value of 480 KA (the primary current is display C and the secondary (load) current is display D on the lower portion of the figure).

Figure 11. Display of the primary current and secondary (load) current during the simulation time.

^

E §

Figure 12. Expanded display of the primary current and secondary (load) current.

Figure 13 shows an expanded display of the current through the opening switch compared with the secondary (load) current. The opening switch is opened at 920.1 msec. Marker 1 is set at 920.1 msec, which shows an opening switch (labeled ZeroSwt) current value of 19.5 KA and a secondary (load) current value of 480 KA (the opening switch current is display F and the secondary (load) current is display D on the lower portion of the figure).

400

200

400

200

Figure 13. An expanded display of the current through the opening switch and secondary (load) current at quench and when opened.

Discussion The analytical results as presented in Figure 13 should be compared to the corresponding curves as formulated conceptually in Figure 8. The current provided by our particular HPPT unit can be adjusted easily to provide the exact current (500 KA) required to zero out the current through the switch. This exercise provides the indication that the opening switch (SW3) can be opened at a very low current. It should be noted that the HPPT unit provides the

initial plasma ignition energy following which the homopolar generator continues to provide the plasma energy.

The energy loss in the primary is due to a coupling coefficient of 0.9, which results in a helium loss in the primary of only 7-8 liters.

In this conceptual design, the opening switch (SW3) is assumed to have a continuous current rating of 500 KA and an opening current rating of 50 KA, or less. The switch actuating mechanism would include auxiliary contacts, which would signal the quench initiation circuitry to initiate the HPPT primary quench following a specific time delay. This time delay would allow the current through SW3 to be at or near zero when the main contacts open. The critical parameter of the switch design is the uncertainty in the time period between the auxiliary contact actuation and the actual opening of the main contacts. This uncertainty (delta time) must be small enough to permit the "plateau" of near zero current through SW3 to span this delta time. For our design, the permissible delta time span is 400 μsec

References [1] M.A. Hilal and E.M.W. Leung. "Energy Compression and Transfer using Hybrid Pulse Power Transformer (HPPT)", presented at the 6th IEEE Pulsed Power Conference at Arlington, Virginia (June, 1987).

[2] M.A. Hilal. "Hybrid Transformer Current-zero Switch", presented at the 6th IEEE Pulsed Power Conference at Arlington, Virginia (June, 1987).

[3] Analog Workbench User Manual, Analog Design Tools, Inc., Apollo Version. February, 1988.

[4] SPICE PLUS User\'s Guide. Analog Design Tools, Inc. Revision 2.1, August, 1987.