U.S. patent application number 09/824431 was filed with the patent office on 2002-10-03 for capacitance rejecting ground fault protecting apparatus and method.
Invention is credited to Jones, Thaddeus M..
Application Number | 20020140432 09/824431 |
Document ID | / |
Family ID | 34808699 |
Filed Date | 2002-10-03 |
United States Patent
Application |
20020140432 |
Kind Code |
A1 |
Jones, Thaddeus M. |
October 3, 2002 |
CAPACITANCE REJECTING GROUND FAULT PROTECTING APPARATUS AND
METHOD
Abstract
A method of controlling a load in response to a ground fault
condition includes measuring a ground fault alternating current
flowing from the load. A real part and an imaginary part of the
ground fault alternating current is ascertained. Electrical power
is removed from the load and/or the ground fault condition is
indicated to a user if a magnitude of the real part exceeds a first
predetermined threshold and/or a magnitude of the imaginary part
exceeds a second predetermined threshold.
Inventors: |
Jones, Thaddeus M.; (Bremen,
IN) |
Correspondence
Address: |
Todd T. Taylor
TAYLOR & AUST, P.C.
P.O. Box 560
142 S. Main St.
Avilla
IN
46710
US
|
Family ID: |
34808699 |
Appl. No.: |
09/824431 |
Filed: |
April 2, 2001 |
Current U.S.
Class: |
324/509 |
Current CPC
Class: |
G01R 31/52 20200101;
H02H 3/337 20130101; G01R 27/18 20130101 |
Class at
Publication: |
324/509 |
International
Class: |
G01R 031/14 |
Claims
What is claimed is:
1. A method of controlling a load in response to a ground fault
condition, said method comprising the steps of: measuring a ground
fault alternating current flowing from the load; ascertaining a
real part and an imaginary part of the ground fault alternating
current; and at least one of removing electrical power from the
load and indicating the ground fault condition to a user if at
least one of: a magnitude of the real part exceeds a first
predetermined threshold; and a magnitude of the imaginary part
exceeds a second predetermined threshold.
2. The method of claim 1, wherein the first predetermined threshold
is such that both personnel and equipment are simultaneously
protected.
3. The method of claim 1, wherein said ascertaining step comprises:
determining a phase angle of the ground fault alternating current
relative to a branch current entering the load; measuring a
magnitude of the ground fault alternating current; and at least one
of: multiplying the magnitude of the ground fault alternating
current by a cosine of the phase angle; and multiplying the
magnitude of the ground fault alternating current by a sine of the
phase angle.
4. The method of claim 3, wherein said determining step comprises:
ascertaining a period of at least one of the ground fault
alternating current, the branch current and a branch voltage;
measuring a time period between a zero-crossing of the ground fault
alternating current and a zero-crossing of the branch current; and
dividing the time period by the period of at least one of the
ground fault alternating current and the branch current.
5. The method of claim 4, wherein said step of measuring a
magnitude of the ground fault alternating current is performed at a
time corresponding to a peak of the ground fault alternating
current.
6. The method of claim 5, wherein the time corresponding to the
peak is approximately one-quarter of the period of at least one of
the ground fault alternating current, the branch current and the
branch voltage after the zero-crossing of the ground fault
alternating current.
7. The method of claim 6, wherein said first predetermined
threshold is approximately 6 milliamperes and said second
predetermined threshold is approximately 30 milliamperes.
8. The method of claim 1, wherein said cosine of the phase angle
and said sine of the phase angle are stored in a common array.
9. The method of claim 1, wherein said measuring and ascertaining
steps are performed under sinusoidal steady state conditions.
10. The method of claim 1, comprising the further step of at least
one of calibrating and adjusting at least one of said first
predetermined threshold and said second predetermined
threshold.
11. The method of claim 10, wherein said step of at least one of
calibrating and adjusting is performed using a single positive
power supply.
12. A ground fault circuit interrupter apparatus, said apparatus
comprising: a ground fault current detector configured for
measuring a ground fault current from a load and producing at least
one ground fault signal indicative thereof, and an automatic
control circuit connected to said ground fault current detector,
said control circuit being configured for: receiving said at least
one ground fault signal; ascertaining a real part and an imaginary
part of the ground fault current dependent upon said at least one
ground fault signal; and at least one of removing electrical power
from the load and indicating a ground fault condition if at least
one of: a magnitude of the real part exceeds a first predetermined
threshold; and a magnitude of the imaginary part exceeds a second
predetermined threshold.
13. The apparatus of claim 12, wherein said ground fault current
detector includes a current transformer connected to the load.
14. The apparatus of claim 13, wherein said current transformer has
an output, said ground fault current detector including a
non-inverting amplifier connected to said output of said current
transformer.
15. The apparatus of claim 12, wherein said at least one ground
fault signal includes a pulse signal indicative of a phase angle of
the ground fault current relative to a branch current entering the
load.
16. The apparatus of claim 15, wherein said pulse signal includes a
plurality of pulses, a duration of each said pulse corresponding to
a time period between a respective zero-crossing of the ground
fault current and a respective zero-crossing of one of the branch
current and a branch voltage.
17. The apparatus of claim 12, wherein said automatic control
circuit includes a switch having a first position and a second
position, said control circuit being configured for removing
electrical power from the load when said switch is in said first
position, said control circuit being configured for indicating a
ground fault condition to a user when said switch is in said second
position.
18. The apparatus of claim 12, wherein said automatic control
circuit includes a contactor configured for removing electrical
power from the load.
19. The apparatus of claim 12, wherein the load comprises a
shielded heater wire.
20. A method of detecting a ground fault condition, said method
comprising the steps of: making a plurality of ground fault current
measurements, each said ground fault current measurement being made
at a respective point in time; calculating an average of said
ground fault current measurements; and comparing said average to a
predetermined threshold value.
Description
BACKGROUND OF THE INVENTION
[0001] 1. Field of the Invention
[0002] The present invention relates to a ground fault protection
apparatus, and, more particularly, to a ground fault protection
apparatus for preventing shock and/or equipment damage.
[0003] 2. Description of the Related Art
[0004] The U.S. National Electrical Code (NEC) requires ground
fault protection for both shock and equipment protection. Although
shock protection requires a 6-milliampere limit, there is no NEC
current limit for equipment protection. In the U.S., a figure of 30
milliamperes is commonly used and 100 milliamperes in Canada.
[0005] The are two types of ground fault protection apparatus. A
ground fault circuit interrupter (GFCI) opens its branch circuit
upon detecting a ground fault current exceeding a maximum limit.
Current cannot be restored to the branch circuit until the GFCI is
manually reset. GFCI applications include residential kitchens,
outdoor applications, and bathroom branch circuits, including those
for floor warming and heating. In residential applications, the
GFCI limit is 6 milliamperes for personnel protection and 30
milliamperes for heating apparatus and equipment protection.
[0006] The second type of ground fault protection apparatus warns
of a ground fault hazard but does not interrupt current flowing in
a branch circuit. Warning is used only in fire protection
applications where the ground current hazard is considered less
dangerous than interrupting the current to pipe trace heaters that
keep wet sprinkler systems from freezing.
[0007] The ground fault current is the vector sum of the currents
flowing in a branch circuit. If there is no ground fault current
flowing, the branch currents sum to zero. In the event of a ground
fault current, the branch currents do not sum to zero. Their
difference is the ground fault current.
[0008] FIG. 1 shows a balanced electrical branch circuit 10 with no
ground fault current flowing. Circuit 10 includes input wiring 12
and a two-pole circuit breaker 14 providing over-current protection
and circuit interruption. Two-pole circuit breaker 14 is required
in electrical systems without a grounded neutral including low
voltage (i.e., less than or equal to 600 volts AC) branch circuits
using U.S. common distribution voltages. These include 240 volts
single and three-phase along with 208 and 480 volts three-phase.
U.S. distribution voltages with a grounded neutral include 120
volts single phase along with 277 volts three-phase.
[0009] A current i.sub.1 flows to a load 16 through a ground fault
protector 18. Similarly, a current i.sub.2 flows from the load 16
through ground fault protector 18. In FIG. 1, no ground current
flows. Thus, the sum of the currents i.sub.1 and i.sub.2 is
zero.
[0010] FIG. 2 shows the case with a ground fault current i.sub.3
flowing from a load 20. FIG. 2 is identical to FIG. 1 except that
the ground current i.sub.3 flows to equipment ground. A branch
circuit 22 must supply the ground fault current i.sub.3. Thus, the
ground fault current i.sub.3 equals the difference between i.sub.1
and i.sub.2.
[0011] The ground fault current i.sub.3, expressed as a vector, has
both magnitude and phase. This is caused by the fact that there is
capacitance between the current-carrying branch circuit 22 and
ground. The reactive, imaginary current component 24 (FIG. 3)
flowing through the capacitance is at a right angle to the
in-phase, resistive, real component 26. Since capacitance is purely
reactive, current flowing through it does not cause heating.
Further, such capacitance is not indicative of a shock hazard.
Thus, capacitance does not indicate a threat to either personnel or
equipment. The resistive component, in contrast, does cause heat
and is indicative of a threat to both personnel and equipment. So
far as fire safety is concerned, only the real current causes
heating. The imaginary component does not.
[0012] In a typical cable configuration heater 28 as is shown in
FIG. 4, a heater wire 30 is surrounded by insulating material 32.
Failure of the heater's insulation 32 causes a substantial in-phase
ground fault current to flow. A shield 34 provides fire safety by
diverting current resulting from insulation or mechanical failure
to the shield 34 which is connected to the safety ground (i.e.,
earth ground). Shield 34 conducts this current to safety ground,
thus providing protection until the GFCI or ground fault protector
18 detects a ground fault current above a threshold value and
interrupts current flow in the branch circuit 22. Thus, the fire
hazard is eliminated.
[0013] Heating cable 28 can be used for pipe trace heating, floor
warming and heating, ceiling and wall heating along with many
industrial applications for process heating. Although cable heaters
employ a wide variety of construction schemes and insulating
material, they all employ a grounded outer braided shield 34 or
stainless steel or copper jacket as required by the NEC. This
construction eliminates the fire hazard that would otherwise occur
if insulation 32 failed for any of a variety of reasons.
[0014] FIG. 5 shows the equivalent lumped circuit of the heater and
the elements causing the flow of the ground fault current i.sub.3.
A substantial capacitance 36 between the heating element 30 and
equipment ground (i.e., safety ground) exists that is proportional
to the heater length. The application of supply voltage to the
heating element 30 causes a substantial current to flow through
this capacitance 36 to equipment ground. This represents a ground
fault current i.sub.3.
[0015] A leakage resistance 38 and heater-to-shield capacitance 36
are shown as acting at the center of the cable heater 28. This
simplification is reasonable since the leakage resistance 38 and
leakage reactance 36 are much greater than the heater resistances
40 and 42. The leakage currents i.sub.4 and i.sub.5 flow into the
equipment ground 44 (i.e., safety ground).
[0016] The vector sum of the currents i.sub.4 and i.sub.5 equal
i.sub.3 which is the ground fault current. From FIG. 5, it is shown
that the ground fault current i.sub.3 has two components: i.sub.4
which is real and i.sub.5 which is imaginary. The real component 4
is in phase with the branch distribution voltage across input
wiring 12. The imaginary component i.sub.5 leads the real component
i.sub.4 by ninety degrees. FIG. 3 shows the vector relationship
between these currents when expressed as phasors.
[0017] The commonly used 30-milliampere GFCI setting for equipment
protection does not eliminate the shock hazard. In heating
applications, the 30-milliampere limit creates both economic and
safety problems. The 30-milliampere GFCI setting limits the length
of heater cable that can be powered by a single branch
circuit--particularly at the higher distribution voltages of 277
and 480 volts (600 volts in Canada). The capacitance 36 between the
shield 34 and the heater wire 30 is proportional to length, as is
the ground fault current. The 30-milliampere setting is too high to
provide shock protection.
[0018] What is needed in the art is a method of identifying the
real and imaginary parts of a ground fault current.
SUMMARY OF THE INVENTION
[0019] The present invention provides a method for providing both
shock and equipment protection in a single GFCI or ground fault
protection device by rejecting or ignoring all or most of the
ground fault current that is due to capacitance between the heaters
and distribution bus wiring and ground.
[0020] The invention comprises, in one form thereof, a method of
controlling a load in response to a ground fault condition. The
method includes measuring a ground fault alternating current
flowing from the load. A real part and an imaginary part of the
ground fault alternating current is ascertained. Electrical power
is removed from the load and/or the ground fault condition is
indicated to a user if a magnitude of the real part exceeds a first
predetermined threshold and/or a magnitude of the imaginary part
exceeds a second predetermined threshold.
[0021] An advantage of the present invention is that it is possible
to consider only the real part of a ground fault current when
determining whether the ground fault current requires a
response.
[0022] Another advantage is that it is practical to simultaneously
provide ground fault protection to both personnel and
equipment.
BRIEF DESCRIPTION OF THE DRAWINGS
[0023] The above-mentioned and other features and advantages of
this invention, and the manner of attaining them, will become more
apparent and the invention will be better understood by reference
to the following description of an embodiment of the invention
taken in conjunction with the accompanying drawings, wherein:
[0024] FIG. 1 is a schematic diagram of a known branch circuit with
no ground fault current;
[0025] FIG. 2 is a schematic diagram of a known branch circuit with
a ground fault current;
[0026] FIG. 3 is a phasor diagram of the ground fault current of
FIG. 2;
[0027] FIG. 4 is a schematic, cross sectional view of a known
heating cable;
[0028] FIG. 5 is a schematic diagram of a simplified equivalent
circuit of the heating cable of FIG. 4;
[0029] FIG. 6a is a plot of the ground fault current of FIG. 2
versus time;
[0030] FIG. 6b is a plot of the voltage across the input wiring of
the branch circuit of FIG. 2 versus time;
[0031] FIG. 6c is a pulse waveform indicative of the time
difference between the ground fault current of FIG. 6a and the
branch voltage of FIG. 6b;
[0032] FIG. 7 is a block diagram of one embodiment of a ground
fault protector of the present invention;
[0033] FIG. 8 is a schematic diagram of the microcontroller
subsystem of FIG. 7; and
[0034] FIG. 9 is a schematic diagram of the power control and
sensing system of FIG. 7.
[0035] Corresponding reference characters indicate corresponding
parts throughout the several views. The exemplification set out
herein illustrates one preferred embodiment of the invention, in
one form, and such exemplification is not to be construed as
limiting the scope of the invention in any manner.
DETAILED DESCRIPTION OF THE INVENTION
[0036] The present invention accurately measures both the real and
imaginary parts of the ground fault current. Although this
measurement can be performed using analog or digital techniques, a
digital method, as described herein, is simpler and lower cost.
[0037] Ideally, rejecting or ignoring the imaginary component
i.sub.5 of the ground fault current i.sub.3 provides superior
protection since the real component i.sub.4 which causes heating is
detected. Furthermore, sensitivity to the real component i.sub.4 is
such that the 6 milliampere personnel safety current limit can be
maintained while providing a 30 milliampere, or higher limit for
the reactive component. Thus, simultaneous equipment and personnel
protection is both practical and possible.
[0038] By assuming sinusoidal steady state conditions, the real
component i.sub.4 and imaginary component i.sub.5 of i.sub.3 can be
trigonometrically calculated from the magnitude of i.sub.3 if the
phase angle .theta. is known. The equations follow:
Re{i.sub.3}=i.sub.3 cos .theta.
Im{i.sub.3}=i.sub.3 sin .theta.
[0039] The phase angle .theta. is determined by measuring the phase
shift between the ground fault current i.sub.3 (FIG. 6a) and the
branch voltage (FIG. 6b) across input wiring 12. This can be
accomplished by measuring the time difference between the positive
going zero crossing of the ground fault current waveform (FIG. 6a)
and the next positive going zero crossing of the branch voltage
waveform (FIG. 6b). The phase angle 0 is calculated using the
following equation: .theta.=360.degree. (time difference)
frequency
[0040] wherein
[0041] `.theta.` is the phase shift in degrees;
[0042] `f` is the power line frequency in Hertz; and
[0043] `time difference` is the time between zero crossings in
seconds.
[0044] The time difference is indicated by a width of each
individual pulse in a pulse waveform (FIG. 6c).
[0045] Using the above procedure eliminates the need for high speed
real-time calculation. Only the time difference needs to be
measured. This is accomplished with a simple time period
measurement that is a built-in function of most microcontrollers.
After determining the time difference, the phase shift angle
.theta. can be easily calculated from the expression show above.
The values of sin .theta. and cos .theta. can be determined from a
look-up table. Simple multiplication yields the values of
Re{i.sub.3} and Im{i.sub.3}.
[0046] The above-described procedure for calculating the values of
Re{i.sub.3} and Im{i.sub.3} is a simple and inexpensive method for
obtaining the desired result without the need for high speed
arithmetic. For example, these calculations can also be performed
real-time using a digital signal processor.
[0047] One embodiment of a ground fault protector 44 (FIG. 7) of
the present invention is shown attached to a load 20. A power
control and sensing subsystem 46 performs the higher level
functions including power control and converting the ground fault
current and branch circuit voltage into signal levels required by a
microcontroller 48 (FIG. 8) of microprocessor subsystem 50.
Microprocessor subsystem 50 performs computational, timing, display
and operator interface tasks.
[0048] Wires 52, 54, 56 and 58 conduct signals between power
control and sensing system 46 and microcontroller subsystem 50.
Operating voltage and ground connections have been omitted for
clarity.
[0049] The branch circuit connections are made through input wiring
12. Load 20 is connected to system 46 through wires 60 and 62.
[0050] Power control and sensing subsystem 46 is capable of sensing
the ground fault current, functionally checking the operation of
ground fault protector 44 and interrupting load current upon
command.
[0051] A control transformer 64 (FIG. 9) reduces the branch voltage
to a convenient value without introducing phase error. Its
secondary voltage provides current for self-testing the ground
fault protection function and the phase reference used to determine
the real and imaginary components of ground fault current
i.sub.3.
[0052] A double pole contactor 66 interrupts current to load 20
during self-test of ground fault protector 44 or in the event the
ground fault current i.sub.3 exceeds a preset value. Both sides of
the branch circuit are broken. This is necessary to interrupt the
ground fault current in power distribution systems without a
grounded neutral, e.g., 240 volts 3-wire, 208 volts 3-phase and 480
volts 3-phase in the U.S.
[0053] Microcontroller 48 placing a logic voltage on input lead 56
to the gate of N-channel metal oxide field effect transistor
NMOSFET 68 causes current to flow through the solenoid coil of
double pole contactor 66. This pulls-in or closes the contacts 70
of contactor 66, thus applying the branch circuit voltage to load
20. The contactor coil operates from the DC supply voltage V++. A
diode 72 protects NMOSFET 68 from a destructive inductive voltage
transient when interrupting current to the contactor coil. A
pull-down resistor 74 prevents spurious contactor operation if the
NMOSFET gate wire 56 is open-circuited, as is often the case during
power-on initialization of associated microcontroller 48.
[0054] A four-winding current transformer 76 performs a variety of
functions. One function is summing the branch circuit currents
flowing through single-tum winding number one 78 and single-turn
winding number two 80. Winding number three 82 is connected to
current shunt resistor 84. This causes a voltage to appear across
shunt resistor 84 that is proportional to its value divided by the
number of turns of wire forming winding number three 82. Winding
number four 86 is an auxiliary winding used for self-test
purposes.
[0055] Microcontroller 48 placing a logic level voltage on input
lead 58 to the gate of NMOSFET 88 causes current to flow through a
coil of a relay 90, which causes the relay's contact to close. This
causes current supplied by control transformer 64 to flow through a
limiting resistor 92 and thence through the self-test winding 86 of
the four-winding current transformer 76. This simulates a ground
fault current above the threshold value.
[0056] The coil of relay 90 is supplied from the V++ voltage. A
diode 94 protects NMOSFET 88 from a destructive inductive voltage
transient when interrupting current to the relay coil. A pull-down
resistor 96 prevents spurious relay operation if the NMOSFET gate
wire 58 is open-circuited, as is often the case during power-on
initialization of associated microcontroller 48.
[0057] An operational amplifier 98 is configured as a non-inverting
amplifier. Its voltage gain is determined by the ratio of its
feedback resistors 100 and 102. This assumes that the value of
resistor 100 is much greater than the value of shunt resistor 84.
Thus, the output voltage of the operational amplifier 98 appearing
at wire 54 is linearly proportional to the magnitude of the ground
fault current.
[0058] An operational amplifier 104 is configured as a
non-inverting amplifier. Its voltage gain is determined by the
ratio of its feedback resistors 106 and 108. The output voltage of
control transformer 64 is buffered and reduced in amplitude by the
voltage amplifier employing operational amplifier 104. The buffered
output appears at wire 52.
[0059] FIG. 8 shows the microcontroller 48, along with support and
interface elements. Support elements include a crystal resonator
110 whose function is to provide a stable accurate clock frequency
for microcontroller 48. This insures the accurate timing functions
required by the invention.
[0060] A supervisor support element 112 insures predictable
start-up of microcontroller 48 upon the application of power.
Supervisor 112 also prevents electrical transients from upsetting
the operation of microcontroller 48. Supervisor 112 asserts
microcontroller restart by holding the microcontroller's RST input
114 high unless the power supply voltage is stable. Supervisor 112
asserts restart input high unless its `watch dog` input 116 is
toggled every 100 milliseconds or so. This prevents microcontroller
48 from latching and thus failing to perform its required
functions. When microcontroller 48 periodically emits the `watch
dog` toggle, it insures that it is properly executing its
program.
[0061] A switch 118 selects the response to a ground fault
condition. The default condition is selected with the switch 118
open as is shown. The second response occurs if the switch 118 is
closed. Resistor 120 provides a pull-up to V+. This provides a
logic level change that is inputted to microcontroller port
122.
[0062] A pushbutton switch 124 toggles the TEST/RESET of the ground
fault protection function. A resistor 126 performs a pull-up
function. Pushing switch 124 provides a logic level change at the
microcontroller input port 128.
[0063] A light emitting diode (LED) ground fault indicator 130
operates while a microcontroller port 132 is logically high. A
resistor 134 sets ground fault indicator 130 to its design
value.
[0064] A potentiometer 136 creates an adjustable bias voltage at an
analog-to-digital converter (ADC) input port 138. This bias sets
the ground fault trip current value for the real part of the ground
fault current i.sub.3.
[0065] Another potentiometer 140 creates an adjustable bias voltage
at an ADC input port 142. This bias sets the ground fault trip
current value for the imaginary (i.e., capacitive) part of the
ground fault current i.sub.3.
[0066] A microcontroller output port 144 provides an output signal
for operating two-pole contactor 66. Similarly, a microcontroller
output port 146 provides an output signal for operating relay
90.
[0067] Comparators 148 and 150 along with the NAND gate 152
generate a pulse proportional to the phase difference between the
branch circuit voltage and the ground fault current i.sub.3. The
pulse (FIG. 6c) is inputted to a microcontroller timer port
154.
[0068] Microcontrollers commonly provide a facility for measuring
the period of an external waveform. This is accomplished by gating
a train of internally generated pulses, derived from the
microcontroller's crystal-controlled clock, into an internal
register. The external signal controls the gate. For example, the
gate could open on the leading edge of the external waveform and
close on the trailing edge. The contents of the register, which is
proportional to the period of the external waveform, can be
transferred to the microcontroller's program counter, or equivalent
register.
[0069] Either of two ground fault protection responses are provided
to a ground fault condition. Switch 118 selects the response. The
default response is to maintain power interruption to the load 20
until the resetting of the ground fault condition, which is
accomplished by pressing switch 124. Leaving switch 118 in its
normally open position selects this response. When selecting this
mode, microcontroller 48 must store the information that a ground
fault occurred in EERAM 156 or its equivalent (i.e., flash RAM).
The U.S. and Canadian NEC require power interruption to the load
after operating power is restored to ground fault protector 44 in
the event of an interruption.
[0070] The second response is to indicate the ground fault
condition while the ground fault condition exists. Power to the
load 20 is not interrupted. No reset action is required if the
condition clears. This response in enabled by closing switch 118
only in certain fire protection applications where the ground fault
condition is a secondary consideration to maintaining load power.
This is the case for heater controls in wet sprinkler systems.
[0071] If no ground fault condition exists, pressing switch 124
automatically verifies proper ground fault protection operation.
Verification consists of a sequence of steps. Immediately after
switch 124 has been pressed, two-pole contactor 66 is de-energized,
if it is energized, thus removing power to load 20. This removes
external ground fault current to insure verification accuracy.
[0072] Next, relay 90 is energized, thus applying the test current
to the current transformer winding four 86. This current simulates
a ground fault current above the real threshold value. The ground
fault indicator 130 will operate for approximately two seconds. If
the ground fault test fails, ground fault indicator 130 will flash
continuously and two-pole contactor 66 will remain de-energized as
is the case with the default response (i.e., when switch 118 is
open).
[0073] If switch 118 is closed, thus selecting the warning mode, a
verification failure is identified by ground fault indicator 130
continuing to flash. However, normal operation of two-pole
contactor 66 will resume.
[0074] If the test is successful, relay 90 is de-energized along
with ground fault indicator 130. Next, two-pole contactor 66
resumes the state that it was in before switch 124 was pressed. A
new test sequence cannot be initiated unless switch 124 has been
released and is not pressed for two contiguous seconds.
[0075] Potentiometers 136 and 140 set the real and imaginary (i.e.,
capacitive) ground fault trip current levels, respectively.
Currents exceeding these levels cause ground fault protector 44 to
operate to remove power from load 20. It is possible to make these
calibrated adjustments accessible to maintenance personnel.
Normally, potentiometers 136, 140 are used to calibrate ground
fault protector 44 during manufacture.
[0076] The voltage between potentiometers 136 and 140 and ground is
linearly proportional to the wiper position. This makes it possible
to calibrate these adjustments. The hardware embodiment herein
described operates from a single positive power supply. The AC
voltages inputted to the microcontroller A-D input ports 138, 142
are half-wave. This reduces analog signal processing circuit
complexity and costs (e.g., elimination of the need for a second
negative power supply along with DC level shifting components).
[0077] The wiper voltages are encoded by the microcontroller's A-D
converter and thereafter stored in specific random access memory
(RAM) locations. The microcontroller inputs 138 and 142 are
serviced by the internal A-D converter.
[0078] Determining the real and imaginary ground fault current
values involves executing a sequence of processes. Conceptually, a
process can viewed as being similar to a subroutine or subprogram.
However, unlike a subroutine, a process can describe a sequence of
steps that can execute as a program. The words "subroutine" and
"process" are used interchangeably herein.
[0079] The first process includes the steps required to determine
the magnitude of the ground fault current. The positive peak value
of the ground fault current waveform (FIG. 6a) is measured since it
is linearly proportional to the magnitude.
[0080] As is shown in FIG. 6a, the ground fault current waveform is
sinusoidal. Thus, its peak value occurs 90 degrees after zero
crossing. This occurs at a point in time that is one-quarter of the
period of the sinusoid after the zero crossing. With a 60 Hz power
line frequency, the peak occurs approximately {fraction (1/240)}
second (0.0041667 second) after the zero crossing. Note that the
derivative of the ground fault current waveform (FIG. 6a) with
respect to time is zero at the 90 degree point. Thus, the amplitude
of the ground fault waveform does not change rapidly with respect
to time at the 90 degree point. A plus or minus one degree change
at 90 degrees results in less than a minus 0.016% change in the
peak value. Further note that one degree of phase shift at 60 Hz is
46.3 microseconds.
[0081] Measurement of the peak value of the ground fault current
waveform connected to an input 158 is accomplished by triggering
the microcontroller's A-D 0.0041667 second after the zero crossing.
Microprocessor 48 provides timing capability for this purpose.
Depending upon the resonator frequency selected, the delay time can
be set with an uncertainty that is less than 40 microseconds. The
A-D encodes the value at its input 158 when triggered. The encoding
time can be up to 100 microseconds depending upon the resonator
frequency. The encoded value is added to the contents of a specific
RAM location.
[0082] Noise (i.e., uncertainty) in the ground fault current
magnitude value could cause spurious ground fault protection
operation. Filtering minimizes uncertainty. The ground fault
current magnitude is filtered by adding the four most recent ground
fault current magnitudes to the specific RAM location cited in the
previous paragraph. After each fourth sample, the contents of this
RAM location is shifted left twice. In effect, this divides the
contents of the ground current magnitude location by four. The
resulting value is taken as the ground fault current magnitude for
another process.
[0083] FIG. 6c shows the pulse the duration of which is
proportional to the phase difference between the branch voltage
waveform (FIG. 6b) and the ground fault current waveform (FIG.
6a).
[0084] At a branch supply of 60 Hz, the pulse width is 43.6
microseconds per degree of phase shift. The pulse is applied to the
input 154 of microcontroller 48.
[0085] Microcontroller 48 provides a facility for measuring the
period of a pulse connected to the input port 154. Microcontroller
48 does this by applying a gated periodic pulse train derived from
its resonator controlled clock into an internal register that is
configured as a counter. The period of the internal pulse train is
less than the time interval of one degree of phase shift at the
branch circuit frequency. For example, a period of less than 40
microseconds is adequate for 60 Hz since this provides a resolution
that is better than one degree. Thus, the pulse train frequency
should exceed 25 KHz.
[0086] The external pulse (FIG. 6c) connected to the input port 154
gates the pulse train supplied to the counter. Counting begins with
the positive leading edge of the pulse and stops upon the trailing
edge of the pulse. The resulting number stored in the counting
register is linearly proportional to the duration of the external
pulse. This number is transferred to a unique RAM location for
storage until needed.
[0087] The scaled values for the both the sine and cosine functions
are stored in a single, common lookup table or array of ninety
contiguous locations in program memory 156. Scaling of these values
simplifies future calculation. The array index, that is, phase
angle, determines the sine or cosine value. The symmetry of these
function eliminates the need to store separate values for the sine
and cosine functions. That is, the single array of ninety
contiguous values is used to determine both sine and cosine
values.
[0088] Next, the array index is calculated from the counter value
stored in a unique RAM location, as was described in the previous
paragraph. This requires integer offsets and rotations of the
counter value. Individual array indexes are required to select the
scaled sine and cosine values which are stored in unique RAM
locations.
[0089] The scaled imaginary value is determined by multiplying the
stored sine and stored ground fault current magnitude values
together and the result is stored in a unique RAM location. The
scaled real value is determined by multiplying the stored cosine
and stored ground fault current magnitude values together and the
result is stored in a unique RAM location.
[0090] The output of the calibrated real ground fault setting
potentiometer 136 is connected to the microcontroller A-D input
138. Microcontroller 48 encodes its value and stores it in a unique
RAM location.
[0091] The output of the calibrated imaginary ground fault setting
potentiometer 140 is connected to the microcontroller A-D input
142. Microcontroller 48 encodes its value and stores it in a unique
RAM location.
[0092] The scaling described above in determining the sine and
cosine of the phase angle assures that the internally stored real
and imaginary ground fault current values match the encoded
internally stored real setting and imaginary ground fault current
calibrations. That is, the settings are accurately calibrated in
engineering units of milliamperes.
[0093] Either or both of two conditions command a ground fault
trip. The first condition is the stored real ground current value
equaling or exceeding the stored real setting. The second condition
is the stored imaginary ground current value equaling or exceeding
the stored imaginary setting value. In the event that either or
both these conditions occur, a ground fault condition exists and a
trip is declared.
[0094] As discussed above, ground fault protector 44 has a choice
of two responses to a ground fault condition. Switch 118 selects
the response.
[0095] While this invention has been described as having a
preferred design, the present invention can be further modified
within the spirit and scope of this disclosure. This application is
therefore intended to cover any variations, uses, or adaptations of
the invention using its general principles. Further, this
application is intended to cover such departures from the present
disclosure as come within known or customary practice in the art to
which this invention pertains and which fall within the limits of
the appended claims.
* * * * *