BACKGROUND

[0001] With the aerospace industry moving to more-electric airborne platforms, more and more power conversion equipment is being used for the conversion between different types of power. Such conversion equipment includes families of DC/DC converters (e.g., 28Vdc/270Vdc, 270Vdc/28Vdc); DC/AC inverters (e.g., 28Vdc/115Vac, 270Vdc/115Vac); AC/DC converters (e.g., 115Vac/28Vdc or 115Vac/270Vdc); AC/AC converters (e.g., 115Vac variable frequency to 115Vac 400Hz); and AC/DC/AC or DC/AC variable speed electrical motor drives.

[0002] An important attributes of airborne electrical equipment is its reliability and availability. Conventional power conversion equipment can experience an internal short circuit (i.e. “fault,” “short,” “overcurrent condition,” or “overcurrent fault”) to its intermediate DC power bus (“DC link”), where a high-voltage current over that which the circuit is designed to accommodate (i.e. “overcurrent” or “shoot through current”) is transmitted through the circuit. This abnormal fault condition can allow for the shoot through current to continue to flow and thus damage the equipment, which may reduce the overall power available to the electric loads on the aircraft. Such electrical faults can then cascade to other equipment in communication with the conversion equipment and DC link.

[0003] The internal short on the DC link can be caused by either an internal converter failure, or by an external stimulus such as high intensity radio frequencies (HIRF) or lighting conditions. The internal short may occur when either one of the DC Link capacitors is shorted or when upper and lower switches of the converter are inadvertently commanded into a state of cross-conduction, i.e., when both switches (transistors) are turned on, thus presenting a short circuit. Such short circuit can cause complete failure of the electrical equipment that is connected to the DC link (i.e. where the equipment is inoperative and replacement is required). Preventing such complete failure would be beneficial for the reliable operation of the electrical systems of the aircraft.

[0004] With reference to FIG. 1 , many AC/AC converters and DC/AC converters 2 use the intermediate DC link energy storage stage 4, which is supported by many parallel aluminum electrolytic capacitors 6. An overcurrent condition, caused by either a short on the output stage 8 or a failed capacitor on the DC link 4, can result in a cascading failure to the input stage 10, DC link 4, and traces 12 on a printed circuit board (PCB) 14.

[0005] One solution to address these overcurrent conditions involves the use of a Positive Temperature Coefficient (PTC) thermistor to act as a resettable fuse. PTC thermistors provide an increase in resistance as the current passing through the thermistor increases, which increase of resistance prevents overcurrent conditions from persisting in the circuit to which the thermistor is attached. The electrical resistance of PTC thermistors are reversible, and thus provide resettable overcurrent protection to the circuit. PTC thermistors are arranged in series in the circuit, and are made of a conductive polymer that increases its resistance as temperature rises based on the amount of current passing through it. After the overcurrent condition is removed, the polymer cools and returns (i.e. “resets) to an original, low-resistance state.

[0006] A typical application of such a PTC thermistor 16 is shown in Error! Reference source not found. 2, showing an inverter system 18, including a DC link capacitor 20, a converter part 22 with a diode 24, and an inverter part 26 with an insulated gate bipolar transistor (IGBT) 28, and arranged between an electric power source (VIN) and an electric load (VOUT). The drawbacks of this overcurrent protection scheme are twofold: 1 ) since all load current flows through the PTC thermistor 16, the PTC thermistor 16 introduces inefficiencies to the system during normal operation by reducing the amount of current flowing through it. Second, during an overcurrent condition, temperature changes to the PTC thermistor 16 that cause an increase in resistance happen relatively slowly compared to the time it can take for a short to occur, such as the time it would take an electrolytic capacitor to fully discharge into an output short. Therefore, the PTC thermistor 16 may not offer adequate overcurrent protection against these fast-occurring overcurrent conditions.

[0007] With reference to FIG. 3, another typical overcurrent protection solution involves use of a current sensing element such as a Hall effect sensor (HES) system 30, including all the same features as FIG. 2, except including a HES 32 in place of the PTC thermistor 16. The HES system 30 is used to directly measure the current flowing through DC link capacitor 20 in real time, and then disconnect the power source when a certain current threshold amount is exceeded. The drawback to this overcurrent protection scheme is that the size of the HES 32 needs to be scaled up considerably as the applied electric load increases. Furthermore, as the size of the HES 32 increases to have a higher current rating, the HES 32 experiences a corresponding decrease in the ability to sense the frequency of current flowing through it.

[0008] Thus, these typical overcurrent protection solutions addressing DC link overcurrent failures introduce inefficiencies to such systems in the form of extra weight (reducing the power per kg of the device), or in the form of line resistance. Furthermore, these sensing elements (i.e. , PTC thermistor 16 and HES 32) may have a slow reaction time (in the order of ps to ms), during which the shoot through current may cascade and cause secondary failures to electrical components in communication with the DC link.

SUMMARY

[0009] A method for protecting a circuit from overcurrent faults is provided. The circuit connects an electric power source to an electric load. The circuit includes an input stage connected to the electric power source; a DC link connected to the input stage, and including a plurality of capacitors connected in parallel to each other; and an output stage connected to the electric load and to the DC link, and in electrical communication with the input stage via the DC link. The method includes connecting an electric current sensor to an instrumented capacitor, the instrumented capacitor being one of the plurality of capacitors; connecting a burden resistor to the electric current sensor; connecting a threshold comparator to the burden resistor, the threshold comparator being configured to determine if output voltage from the burden resistor exceeds a reference voltage; and connecting a control unit to the input stage, to the output stage, and to the threshold comparator. The control unit disconnects the input stage from the DC link when the output voltage exceeds the reference voltage.

[0010] An electric power converter includes an input stage configured to be connected to an electric power source that delivers electric power to the input stage; a DC link connected to the input stage, and including a plurality of capacitors connected in parallel to each other; an output stage connected to the DC link, in electrical communication with the input stage via the DC link, and configured to be electrically connected to an electric load for delivery of the electric power to the electric load; an electric current sensor connected to an instrumented capacitor, the instrumented capacitor being one of the plurality of capacitors; a burden resistor connected to the electric current sensor; a threshold comparator connected to the burden resistor, and configured to determine if output voltage from the burden resistor exceeds a reference voltage; and a control unit in communication with the input stage, the output stage and the threshold comparator, and configured to disconnect the DC link from the input stage when the output voltage exceeds the reference voltage.

BRIEF DESCRIPTION OF THE DRAWINGS [0011] FIG. 1 is a schematic diagram of a prior art AC/AC converter with a DC link.

[0012] FIG. 2 is a schematic diagram of a prior art overcurrent protection circuit using a Positive Temperature Coefficient thermistor.

[0013] FIG. 3 is a schematic diagram of a prior art overcurrent protection circuit using a Hall effect sensor.

[0014] FIG. 4 is a schematic diagram of an overcurrent protection system according to the present subject matter.

[0015] FIG. 5 is a graph showing test results of an overcurrent protection system according to the present subject matter.

[0016] FIG. 6 is a graph showing other test results of an overcurrent protection system according to the present subject matter.

[0017] FIG. 7 is a graph showing still other test results of an overcurrent protection system according to the present subject matter.

DETAILED DESCRIPTION

[0018] The present invention improves on two currently used methods of protecting converter DC links, which employ either a HES based current monitoring and overcurrent protection strategy (FIG. 3), or a temperature-based overcurrent protection strategy like PTC thermistors (FIG. 2). These current systems are deficient because they lack the ability to operate to protect the circuit in less than 200 ns, which is the maximum time that most semiconductors can tolerate shoot through currents of high voltage. As such, these systems do not necessarily prevent the complete failure of electrical components in communication with the DC link due to the overcurrent condition.

[0019] The present invention may be used in power conversion equipment that employs an intermediate DC link. In comparison to existing solutions, the present invention provides a fast response to an overcurrent condition occurring in the DC- Link of the electric power converter. The fast-acting DC link protects against overcurrent faults, which can normally irreversibly damage electrical equipment in electrical connection with the DC link. However, since the present invention quickly electrically disconnects the power source, this electrical equipment can be prevented from experiencing complete failure due to the overcurrent faults.

[0020] The invention can also be tuned to a desired level of sensitivity, so as to provide a desired level of overcurrent protection. This may avoid nuisance trips of the system (i.e. electrical disconnection of the power source) during normal operating conditions, e.g. where the system experiences an increase in current, but not one that is an overcurrent that is detrimental to the system. When the sensitivity is so tuned, the system may not undesirably electrically disconnect the power source when the increased current, but not an overcurrent, flows through the system.

[0021] Referring to the figures, a block diagram of an inventive circuit/system 40 is shown in FIG. 4, which may be included as part of an electric power converter. The system 40 employs an electric current sensor 34 electrically connected in series with a lead 36 of one capacitor in the DC link 38. This one capacitor that includes the electric current sensor 34 attached thereto, is herein referred to as the “instrumented capacitor” in order to distinguish it from the other capacitors in the DC link 38. By this configuration, the system 40 employs a rather small electric current sensor 34 that only senses current flowing through the instrumented capacitor, instead of using a large current sensing device to measure all of the current flowing through the entire DC link 38.

[0022] The system 40 includes an input stage 42 where electric power is delivered to the system 40 by an electric power source 44. The input stage 42 is electrically connected to the DC link 38, which is electrically connected to an output stage 46, which delivers electric power to an electric load 48. A control unit 50 is in communication with, and controls, the input stage 42 and the output stage 46. The DC link 38 includes several capacitors electrically connected in parallel to each other, the leads of which are shown in FIG. 4. One of the plurality of capacitors is the instrumented capacitor having the current sensor 34 attached to its lead 36. All of the capacitors of the DC link 38 may be electrolytic capacitors.

[0023] A secondary winding of the current sensor 34 is equipped with a burden resistor 52, which provides an output voltage that is proportional to the current supplied from the current sensor 34.

[0024] A threshold comparator 54 is used to compare the output voltage from the burden resistor 52 against a reference voltage, which can be set to a desired voltage by a user. If the output voltage from the burden resistor 52 is greater than the reference voltage, then the control unit 50 operates to electrically disconnect the input stage 42 from the DC link 38 by issuing a fault command to the input stage 42. The moment at which the output voltage from the burden resistor 52 exceeds the reference voltage is referred to as the “trip point,” which is the point at which the control unit 50 issues the fault command. The control unit 50 may also issue the fault command to the output stage 46, which may operate to disable any switches (not shown) in the output stage 46 to electrically disconnect the output stage 46 from the DC link 38. The control unit 50 may also, or alternatively, electrically disconnect the input stage 42 from the power source 44, and/or electrically disconnect the output stage 46 from the load 48.

[0025] The current sensor 34 may include, but is not limited to a current transformer (CT), a shunt sensor, or high frequency HES. Other current sensors 34 may be used. The input and output stages may include high performance MOSFET or IGBT semiconductor devices. The control unit may include basic electronic elements, operational amplifiers and comparators.

[0026] The response time for the system 40 to issue the fault command, and thus electrically disconnect the DC link 38 from the input stage 42, is only limited by the propagation delays in the circuitry of the threshold comparator 54 and in the control unit 50. The amount of delay in response time is a function of trace inductance between the location of the fault and the instrumented capacitor, as well as the primary inductance of the current sensor 34.

[0027] In the following equations, C1/R1 are the positive rail bulk storage capacitance of the DC link 38, C2/R2 are the negative rail bulk storage capacitance of the DC link 38, C3/R3 are the instrumented rail capacitor, L3 is the trace inductance between the bulk storage and the instrumented capacitor, and Lp is the primary inductance of the current sensor 34.

[0028] For simplicity of analysis and in the following equations, the equivalent series resistance (ESR) of the capacitors in the DC link 38 are considered negligible for calculation of initial conditions, and the inductances of the trace/current sensor 34 are combined into a single inductance.

[0029] For calculating node voltages, the following Equation 1 and Equation 2 are used:

(Equation 1 )

(Equation 2)

[0030] In Equations 1 and 2, the following equivalent values apply:

7 _Ci = —sC ₁ V _A

Ic3 — ^s CsV _B

[0031] By substituting these equivalents into Equations 1 and 2, the flowing are derived:

(Equation 3) (Equation 4) [0032] Assuming there are zero initial currents prior to a short circuit of the system, the equation is solved as:

(Equation 5)

[0033] The sensing voltage is defined as:

(Equation 6)

[0034] Using a 10:1 current transformer as the current sensor 34 and a 10W burden resistor 52, the following is calculated: (Equation 7)

[0035] Example results for the response time for an inventive circuit 40 to electrically disconnect the DC link 38 from the input stage 42 are shown in FIGS.

5-7. [0036] FIG. 5 shows an inventive circuit response to a 50mQ shoot through. FIG. 6 shows an inventive circuit response to a 200 mQ shoot through. FIG. 7 shows an inventive circuit response to a 1 .5 W shoot through. In each of FIGS. 5- 7 the line with diamonds (¨) represents the input current, and the line with circles (·) represents the DC rail voltage.

[0037] In the examples, 500nFI was used as the primary inductance of the current transformer as the current sensor 34, and a threshold comparator 54 having a 10ns propagation delay was used. The shoot through current was applied to the circuit 40 for 1 .5 ps using 50 mQ and 200 mQ of current. FIG. 7 shows the behavior for a 1.5W shoot through current, corresponding to an output current below the established reference voltage.

[0038] The total propagation delay from application of the overcurrent to the electrical disconnection of the input stage 42 from the DC link 38 varied from 20 ns to 200 ns, depending on the impedance of the short, where higher currents resulted in faster issuance of the fault commands and thus faster trips of the system 40. The distance between the instrumented capacitor and the rest of the DC link 38 were varied, and showed that longer distances between the instrumented capacitor and the rest of the DC link 38 resulted in longer propagation delays, and thus a slower issuance of the fault commands and slower trips of the system 40. Furthermore, such response time was independent of the ESR of the capacitors in the DC link 38. This independence of the response time with respect to the ESR of the capacitors in the DC link 38, allows for the use of a range of commercial capacitors in the DC link 38, e.g. electrolytic capacitors, and thus does not require the use of a specific capacitor.

[0039] The system 40 allows for tuning of the trip point by changing the reference voltage used by the threshold comparator 54. This can be accomplished by a resistor value change. Such tuning may be accomplished by programming the threshold comparator 54 to use a desired voltage as the reference voltage. If the reference voltage is increased or decreased by a user, then the threshold comparator 54 may trip the system 40 at a higher voltage or a lower voltage, respectively. The trip point can be fine-tuned based on any of a number of considerations, including the desired application in which the system 40 is employed. [0040] This response time of the circuit 40 for issuing the fault command is about an order of magnitude faster than a conventional HES protection system provides, and a few orders of magnitude faster than that provided by PTC protection systems 18 (FIG. 2). As a result, the fault protection provided by the system 40 can be used to detect a cross-conduction (shoot-through) of semiconductor switches, and provide overcurrent protection in time to prevent these switches from experiencing a complete failure, which requires replacement of the switches.

[0041] The present subject matter includes a method for protecting a circuit from overcurrent faults. The circuit includes the input stage 42 electrically connected to the electric power source 44. The circuit includes the DC link 38 electrically connected to the input stage 42. The DC link includes a plurality of capacitors electrically connected in parallel to each other. The circuit includes an output stage 46 electrically connected to the electric load 48 and to the DC link 38, and in electrical communication with the input stage 42 via the DC link 38.

[0042] The method includes operatively connecting various components in order to arrive at the inventive circuit 40. This is accomplished by electrically connecting the electric current sensor 34 to an instrumented capacitor, the instrumented capacitor being one of the plurality of capacitors in the DC link 38. The burden resistor 52 is electrically connected to the electric current sensor 34. The threshold comparator 54 is electrically connected to the burden resistor 52. The threshold comparator 54 is configured to determine if output voltage from the burden resistor 52 exceeds a reference voltage. The control unit 50 is in connected to the input stage 42, to the output stage 46, and to the threshold comparator 54 in order to communication with these components. The control unit 50 electrically disconnects the input stage 42 from the DC link 38 when the output voltage exceeds the reference voltage.

[0043] The method may include establishing the reference voltage, such as by programming the reference voltage in the threshold comparator 54. The reference voltage may be a desired voltage of a user or a predetermined voltage of a manufacturer of the circuit 40. If the output voltage from the burden resistor 52 exceeds the reference voltage, the threshold comparator 54 may communicate this occurrence to the control unit 50, which may then issue the fault command. The fault command issued by the control unit 50 may cause an electrical switch in the output stage to change from a conducting state where current passes through the electrical switch to a non-conducting state where current does not pass through the electrical switch, thus electrically disconnecting the electric load 48 from the DC link 38. Such electrical disconnection may occur within 20-200 ns from a time the output voltage exceeds the reference voltage.

[0044] It will be appreciated that various of the above-disclosed embodiments and other features and functions, or alternatives or varieties thereof, may be desirably combined into many other different systems or applications. Also that various presently unforeseen or unanticipated alternatives, modifications, variations or improvements therein may be subsequently made by those skilled in the art which are also intended to be encompassed by the following claims.