CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit under 35 USC §119 of the filing date of US Provisional Patent Application Serial No. 62/166,924 filed on 27 May 2015. The entire disclosure of this prior application is incorporated herein by this reference.

TECHNICAL FIELD

[0002] The present disclosure relates generally to energy conversion machines such as motors and generators, and, in an embodiment described herein, more particularly provides a system and method for switching inductor

configurations in multi-port converters.

BACKGROUND

[0003] Electric motors have been developed that utilize multiple inductors or windings (i.e. inductors), that are connected to a power source by a traditional half-bridge power stage, consisting of two source-drain-connected power MOSFETs, or emitter-collector-connected Insulated-Gate

Bipolar Transistors (IGBTs), to connect one end of the inductor to either positive or negative power rail (or leg). Some applications connect both ends of the inductor to the half-bridge, which is commonly described as a full-bridge or H-bridge configuration. Each rail can be individually connected to separate ones of the inductor ends, or

synchronized in operation with other rails in multi-phase systems .

[0004] Furthermore, traditional power converters can utilize Pulse Width Modulation (PWM) technology to match power input and power output requirements , and control current flow through an inductor (or coil) in either

continuous (i.e. buck) or discontinuous (i.e. boost) mode. The input voltage range, output load, and input-output ratio can determine the duty cycle range available for the power converter, which often limits the specified input voltage range to a 2:1 or even 3:1 ratio, for example 85Vac to

240Vac, or 36Vdc-72Vdc.

[0005] While control techniques such as dithering and constant-off-time schemes can positively impact

electromagnetic interference issues and filtering

requirements to meet FCC regulations, Pulse Width Modulation (PWM) machines have a preferred duty cycle range for most efficient operation.

SUMMARY

[0006] The present disclosure provides a matrix of power switches, which can direct current flow through any or a plurality of inductors that are connected in the same power conversion system, by connecting each inductor in at least four ways: to the positive power supply, to the negative power supply, to an upstream inductor, or to a downstream inductor. As used herein, "upstream" and

"downstream" inductors refer to inductors that are configured adjacent (or as neighbors) to the inductor being configured. To connect the inductor being configured to inductors in addition to the upstream and downstream

inductors, the inductor being configured is connected through either one of the upstream and downstream inductors to the additional inductors. It should be understood that there can be multiple upstream and multiple downstream inductors, with the matrix controlling connection of the inductor being configured to the desired upstream or

downstream inductors.

[0007] With the introduction of bi-directional power switches, a multi-port configuration can be realized that effectively blocks current flow in any direction, but can also allow current flow in any direction, as determined by the application and controlled by a pair of common gate- drivers . Some examples of bi-directional power switches are a pair of anti-parallel diode-connected IGBTs, back-to-back connected (i.e. source connected) power MOSFETs, and back- to-back connected High Electron Mobility Transistor (HEMT) power switches based on Sic or GaN semiconductor material. The multi-port configuration can be seen as the matrix arrangement of power switches that enables a multitude of serial/parallel inductor configurations, while substantially maintaining a time constant inductance (L ) /resistance(R) ratio of each serial string regardless of a selected

configuration. Maintaining system control loop stability can be especially important for current control applications that are used for accurate torque control of an electrical machine .

[0008] For example, a 4-inductor arrangement can be configured so that initially all four inductors are

connected in parallel (4x1 configuration), impressing the full supply voltage across each inductor for maximum di/dt response. Once the pre-determined current maximum has been reached, the inductors may be configured in a 2x2

configuration, whereas two serial strings of two inductors are connected in parallel, or serial configuration (1x4), where all four inductors can be connected in series, lowering the maximum current to half or one quarter of the maximum current possible without further PWM techniques.

[0009] It can also be appreciated that a serial connection of three inductors (1x3) for a 4-inductor assembly can be a benefit, whereas one inductor can be disengaged from current conduction, thus reducing thermal stress on the 4-inductor assembly by alternating which of the four inductors in the 4-inductor assembly is disengaged from current conduction within the array of four inductors.

[0010] Similar to the firing order of a rotary engine, which has an odd number of cylinders, the present disclosure may use an odd number of inductors, such as 3, 5, 7, 9, 11, etc., in each plane of the electric motor, so that a power conversion system can take advantage of a well distributed thermal load across all inductors. Such "firing" order could be: 1-3-5-2-4-1 (see FIGS. 7A-7B), or in a two-plane, 5-inductor configuration which uses even numbered inductors in the second plane: 1-2-3-4-5-6-7-8-9-10-1 (see FIGS. 7C- 7D). Such configuration will also reduce the amount of noise through vibration of the mechanical system, as inductors are interacting with a magnetic field with equally balanced timing and force applied on nearly opposing elements of the rotating assembly, thereby limiting a prolonged pull/push on the same side of the rotor axis.

[0011] Unlike conventional half-bridge configurations, which may rely on a current injection point between each inductor directly connected to an adjacent inductor on each connector side, the present system can completely bypass and isolate one inductor, for example, during a fault condition, or routine replacement for maintenance.

[0012] For a DC power application, neither positive or negative power supply will reverse and therefore uni- directional power switches may be used, with a total of 12 power switches : four power switches connected in a typical H-Bridge configuration and four back-to-back connected power switches ( 8 individual power switches ) , connecting each inductor node to either an upstream or a downstream inductor node. For an AC power application, a total of 16 power switches can be used, and configured to include bidirectional switches in the H-bridge.

[0013] Such an integrated matrix of power switches can be constructed as a module assembly using conventional discrete or hybrid manufacturing processes. The power switches can be wire-bonded, sintered, and/or soldered, and combined on a single substrate such as substrates used for the manufacture of power semiconductor devices, including but not limited to Sic, GaN, DMOS, or other MOSFET processes available in the industry. The matrix may switch on a subset of the power switches for DC, AC, or 3-phase AC configurations, which can spread heat distribution across the entire substrate, thereby lowering thermal resistance and improving thermal performance of the module assembly.

[0014] Those familiar with pros and cons of PWM control techniques may appreciate the flexibility of the principles of the present disclosure to minimize, if not eliminate, PWM control issues and reduce switching losses. Also, PWM switching could still be utilized if, for example, an all- in-series inductor configuration still exceeds the desired current setting. Since inductive effects are virtually insignificant during each low frequency switching

transition, a machine can operate at full power on safe DC voltages (less than 48 VDC), such as a single cell battery, as long as sufficient current is provide by the power source .

[0015] While traditional designs favor high-voltage due to a favorable reduction of power losses, traditional motors may exhibit large back-EMF (electro-motive force), which demands higher voltages to overcome the same. At the same time, during start-up, traditional motors, due to the lack of back-EMF, can suffer very large inrush currents that need to be limited either by a temporary start-up resistor, or electronic limits, such as PWM circuits. Lorentz-Force Motors, which observe very low back-EMF, do not require high-voltage to overcome the back-EMF, and therefore have lower inrush currents.

[0016] Furthermore, since a Lorentz Force Motor is a DC machine, and interprets AC power as a time-limited momentary DC power at any given time, current at any time is directly converted into force, yielding much improved efficiencies as opposed to commutating DC motors, brush or electronically commutated AC motors, synchronous AC motors, or Switched Reluctance Machines. The principles of the present

disclosure allow AC power to be connected to this DC machine without significant repercussions. At a zero-crossing (or at least immediately after the zero-crossing) of a current signal on a respective power rail, the power switch matrix can reconnect a positive inductor input from a positive power rail to the negative power rail, from an upstream inductor connection to a downstream inductor connection, from a power rail connection to an upstream or downstream inductor connection, etc., therefore allowing proper current flow and direction to be maintained. As used here in, "at zero-crossing" refers to a point in an oscillating current signal when the current flow is zero as the current signal changes polarity going from either positive or negative polarity to negative or positive polarity, respectively. This point can include immediately before or immediately after the actual point of transition between the current signal's polarity. It is preferred to transition between polarities as close to a no-current-flow position in the current signal's waveform.

[0017] Operators that have operated AC or DC motors in poorly regulated power systems, such as may be found in third world countries, on islands, or in aircraft or maritime vessels, etc. will appreciate that a universal power input can provide a high degree of failure resilience for electrical machines, such as during brown-outs, power source transitions from land to vessel, and/or unregulated power transients.

[0018] These and other features, advantages and benefits will become apparent to one of ordinary skill in the art upon careful consideration of the detailed

description of representative examples below and the accompanying drawings, in which similar elements are indicated in the various figures using the same reference numbers .

BRIEF DESCRIPTION OF THE DRAWINGS

[0019] FIG. 1 is a representative block diagram of an electric motor (rotor not shown), with four Intelligent Power Modules (IPMs) which can embody principles of this disclosure.

[0020] FIG. 2 is a representative block diagram of input/output connections for an IPM of FIG. 1.

[0021] FIG. 3 is a representative block diagram of an IPM Of FIG. 1

[0022] FIG. 4 is a representative block diagram of an inductor drive unit of the electric motor.

[0023] FIG. 5A is a representative block diagram of a power switch matrix for DC only machines .

[0024] FIG. 5B is a representative block diagram of a power switch matrix for single phase AC or DC machines.

[0025] FIG. 5C is a representative block diagram of a power switch matrix for three-phase AC machines .

[0026] FIG. 5D is a table of possible DC inductor configurations for a power switch matrix with a 4-inductor assembly.

[0027] FIGS. 6A-F are representative block diagrams of possible DC inductor configurations of an AC/DC power switch matrix with a 4-inductor assembly.

[0028] FIG. 7A is a representative block diagram of a single plane configuration of three IPMs.

[0029] FIG. 7B is a representative block diagram of a single plane configuration of five IPMs.

[0030] FIG. 7C is a representative block diagram of a dual plane configuration of five IPMs per plane. [0031] FIG. 7D is another representative block diagram of a dual plane configuration FIG. 7C.

DETAILED DESCRIPTION

[0032] FIG. 1 shows an energy conversion machine 100, that can be a motor, generator, actuator, voice coil, induction coil, or similar electromagnetic devices that use inductors (such as windings, coils, etc.) to convert between electrical and mechanical energies (e.g. mechanical energy can be force and/or motion). The machine 100 can include one or more digital signal processors (DSPs) 300 which can communicate to one or more personal computers (PCs) 124 over a command and control network connection 122. A PC 124 can be any device that transmits /receives data and control information to/from the DSP 300 for controlling the energy conversion performed by the machine 100.

[0033] The machine 100 can further include multiple Intelligent Power Modules ("IPMs") 200, 202, 204, 206 that can be connected to the DSP via an internal communications bus 120, which transfers data and control information between the DSP 300 and individual IPMs. Each IPM 200, 202, 204, 206 can be connected to positive and negative power rails as well as adjacent (upstream and downstream)

inductors 400, 402, 404, 406. In the case where the machine 100 is a motor, a power supply 132 can be connected to the positive and negative power rails 130, 131 via a power cable 134. If the machine 100 is a generator, then the power supply 132 could be a power load connected to the positive and negative rails 130, 131 via the power cable 134.

[0034] The IPMs may be connected in a ring-like

configuration with adjacent IPMs connected to each other as shown in FIG. 1 where an upstream connection 152 of one IPM 200, 202, 204, 206 is connected to a downstream connection 154 of an adjacent IPM 200, 202, 204, 206, via the

connection 156. It should be understood that other

configurations of IPMs are possible, such as parallel, series, cubic, star, multiple ring, etc. In any of these configurations, each IPM can have a connection to the positive power rail 130, the negative power rail 131, an upstream connection 152, and a downstream connection 154. Also, each IPM can be connected to an inductor 400, 402, 404, 406. The DSP 300 can use the IPMs to configure the flow of current through each inductor by selectively

connecting the inductor 400, 402, 404, 406 between a pair selected from a group consisting of the positive power rail 130, the negative power rail 131, upstream connection 152, downstream connection 154, or open circuit connection. It should be understood, that an open circuit connection to the inductor can be used to prevent current flow through the inductor 400, 402, 404, 406. As used herein, "an open circuit connection" refers to an absence of an electrical connection to an end 146, 148 of the inductors 400, 402, 404, 406. Therefore, when the power switch matrix connects one of the ends 146, 148 to "an open circuit connection", it means that the end connected to the open circuit connection is not connected to any of positive or negative power rails or the upstream or downstream connections .

[0035] As an example of reconfiguring an inductor configuration, the inductor group of FIG. 1 can be

configured by the DSP 300 into a series connection with all four inductors 400, 402, 404, 406 connected in series between the negative power rail 131 to the positive power rail 130. To construct this inductor configuration, the DSP 300 may command the IPM 200 to connect the inductor 400 between the negative power rail 131 and the downstream connection 154. The DSP 300 may command the IPM 202 to connect the inductor 402 between the upstream connection 152 of the IPM 202 and the downstream connection 154 of the IPM 202. The DSP 300 may also command the IPM 204 to connect the inductor 404 between the upstream connection 152 of the IPM 204 and the downstream connection 154 of the IPM 204. The DSP 300 may also command the IPM 206 to connect the inductor 406 between the upstream connection 152 of the IPM 206 and the positive power rail 130. Therefore, in this series configuration of the inductors 400, 402, 404, 406, current would be directed from the positive power rail 130, through the inductor 406, through the inductor 404, through the inductor 402, and then through the inductor 400 to the negative power rail 131.

[0036] To modify the configuration of the inductors 400, 402, 404, 406, such as when a series configuration is not desired, then the DSP 300 can merely command the IPMs to connect their respective inductors 400, 402, 404, 406

between the various connections (e.g. power rails, upstream and downstream IPMs, open circuit connection) to achieve the new configuration. For example, if it is desired to connect all the inductors in a parallel configuration, then the DSP 300 can command the IPMs to connect their respective

inductors 400, 402, 404, 406 between the positive and negative power rails 130, 131, thereby providing a parallel current path through each inductor between the power rails 130, 131. It should be understood that each inductor 400, 402, 404, 406 can comprise multiple inductors. Therefore, references to any one of the inductors 400, 402, 404, 406 can refer to one or more inductors, which can be connected in parallel and/or series within the referenced inductor. [0037] The IPMs 200, 202, 204, 206 provide a matrix of switches that can be used to configure the connection of the inductors 400, 402, 404, 406 in real-time such that the current flow through the inductors is maintained in the correct direction to convert energy from electrical to mechanical, or visa versa. For example, in the case of an AC motor, the input power is continually changing it's polarity. With this change in polarity, the current through an inductor connected between the positive and negative power rails would also continually change direction, which could continually change the direction of a rotational force applied to a rotor of the AC motor. This continual

direction change would not allow the rotor to turn, thereby rendering the motor useless. However, with the

configurability of the inductor connections via a matrix of power switches in each IPM, the connections to the inductors can be continuously changed such that the flow of current through the inductors is maintained in one direction, even though the amplitude of the current through each inductor can change.

[0038] The matrix of power switches in each individual IPM 200, 202, 204, 206 forms a power switch matrix 210 in each IPM, where the matrix 210 can be quickly reconfigured to accommodate the speed at which the inductor connections are changed to prevent reversal of current in any of the inductors 400, 402, 404, 406. The DSP 300 controls the configuration and reconfiguration of the matrix by

commanding the individual IPMs 200, 202, 204, 206 to connect the respective inductors 400, 402, 404, 406 as desired. The DSP 300 can use specific Look-Up Tables (LUTs) to provide switch control for the switch matrix 210. Each LUT can contain an array of switch commands that have been

determined to provide the most efficient and/or most powerful configuration, the best switching sequence, etc., for a plurality of bi-directional power switches contained in each IPM 200, 202, 204, 206.

[0039] It should be understood that inductor

configurations can be adjusted and/or altered based on sensor readings received by the DSP 300 from one or more feedback sensors 310. The sensors 310 can provide sensed data to the DSP 300 related to internal temperatures, substrate temperatures, current flow, voltage potential, rotor speeds, phase angles, torque, strain, speed, position, etc. to allow the DSP to make various adjustments, such as preventing current flow through various inductors 400, 402, 404, 406 to minimize heat generated by current flow through the respective inductors, or configuring the inductors 400, 402, 404, 406 to achieve a stable operating point for the machine .

[0040] Accordingly, the machine 100 can isolate and configure any number of inductors in any serial, parallel, or serial/parallel configuration, for example allowing current to selectively flow in either direction through each inductor 400, 402, 404, 406, or combinations thereof. The machine 100 may also contain multiple bi-directional

upstream and downstream power switches (see FIGS. 4-5C), such as a back-to-back configured power MOSFET, IGBT, Sic, or GaN (Gallium Nitrite) power switches. The machine 100 can be powered from a power source 132 (e.g. AC, DC,

multiple-phase AC, etc.) connected through the power cable 134 to the machine's power terminals 140 DC/AC (see FIG. 2) or 160 AC (see FIG. 5C) to an internal power bus (power rails 130, 131 for single-phase, or power rails 162, 164, 166 for three-phase). Each IPM 200, 202, 204, 206 can connect it's respective inductor 400, 402, 404, 406 to either the positive power supply terminal 142, negative power supply terminal 144 (see FIG. 5A) , hot phase 172, return 174 (see FIG. 5B), or a hot phase or return of one of up to three phases 162, 164, or 166 (see FIG. 5C), as well as to upstream and down stream inductors .

[0041] At any given time, the DSP 300 can determine a preferred inductor configuration by calculating an amount of current that can be forced through each inductor 400, 402, 404, 406, depending on its inductance, inductor resistance, and available supply voltage. The DSP can receive such data from each IPM via an internal communication bus 120 (e.g. CAN, LIN, another serial bus), and can calculate the

required current from a user input received through a Human Machine Interface 124 (e.g. PC, tablet, smartphone,

touchscreen device, keypad, etc.) connecting to the energy conversion system with wired (e.g. Ethernet, USB, CAN, Profibus, etc.) or wireless communication (e.g. 802.15,

Zigbee, Bluetooth, Weightless, etc.) via a communication link 122. The DSP 300 can compare such user input with its primary feedback sensor 310 output (e.g. torque, strain, speed, position, etc.), and can calculate a rate of change required to reach a stable operating point for the machine 100.

[0042] FIG. 2 shows an IPM, which can include a power switch matrix 210, power regulator block 220, micro- controller unit (MCU) 230, internal temperature sensor 240, a provision 246 for attaching an external thermo-coupler to measure the internal inductor or winding temperature, analog current sensing 250 and analog mixed-signal block 260. It should however be understood that the IPM is not required to include each of these items. The power terminal block 140 can connect each IPM with its neighboring upstream and downstream IPMs, and can provide for a second, independent communication port 126 for test, debugging, and firmware upgrades, as well as a visual fault indicator output 128, such as an LED, and ground reference connections for both analog 136 and digital 138 ground. The MCU 230 can

communicate with the DSP 300 over the communication link 120 to transfer data and command information to/from the DSP 300. A positive power rail can be connected to connection 142, with a negative power rail connected to connection 144. First and second ends of an inductor 400, 402, 404, 406 can be connected to connections 146, 148 of its respective IPM 200, 202, 204, 206.

[0043] FIG. 3 further details the IPM, which can also include a mixed-signal analog block 260 configured as a programmable current limit, including an analog-digital- converter (ADC) 232, digital-analog-converter (DAC) 267, and multiplexer (MUX) 234, which can be internal or external to the MCU 230. The MCU 230 can include an internal

temperature sensor 240 for measuring its ambient

temperature, and/or utilizing external digital temperature sensors, as well as a temperature sensor 242 dedicated to measuring the temperature of the power switch matrix 210, and thermocouple 244 to measure winding temperature. The internal power regulator 220 can include one or more DC/DC converters 222, 226 and linear regulator (s) 236, to provide internal system voltages, such as, but not limited to:

a) a regulated gate drive voltage via gate drive bus

224,

b) an intermediary power bus 228, and

c) post-regulated analog (e.g. 5V), digital (e.g.

3.3V), and micro-processor and DSP core voltages (e.g.

1.8V).

[0044] Referring now to FIG. 4, each IPM 200, 202, 204, 206 can monitor its power return current (144) via a current shunt sensor 250, processed by instrumentation amplifier 262 and converted to digital with the analog-digital converter (ADC) 264, which may be integrated into the MCU 230, or be an external device which may provide for higher speed or higher resolution. This signal can be compared to a

predetermined set point by the MCU 230, by applying logic algorithms to determine when a transition to a different inductor configuration is required. The MCU 230 can average output current from the selected inductor configurations as required by the requested torque setting, which uses the motor's momentary torque constant kT to determine the inductor current that's needed.

[0045] At very low current settings, the motor may also utilize conventional pulse-width modulation (PWM)

techniques. However, the control method of this disclosure substantially lowers switching losses and electro-magnetic interference as the transition from one switch configuration to another can occur in milliseconds and/or larger time intervals, compared to the high speed PWM switching schemes in sub-millisecond to microsecond time intervals. With the advancement of GaN power transistors, conventional highspeed PWM techniques can take advantage of the matrix configurations disclosed herein, as switching losses are reduced significantly.

[0046] To determine the total system current in all inductors, each return current from each IPM 200, 202, 204, 206, as collected from each MCU 230, can be summed by the central DSP 300, compared to a desired value, and then corrected with a control algorithm (e.g. feedback, feed- forward, PID, PI, fuzzy logic, etc.).

[0047] The analog output signal is also compared by a Schmitt Trigger latch 268 with a current limit signal that can be compensated for temperature provided by the external winding thermo-coupler 244 and/or power stage temperature sensor 242. A current limit threshold can be programmed into the digital-analog converter (DAC) 266, which may be integrated into the MCU 230, or provided as discrete

components .

[0048] Exceeding the current limit can set a high- priority interrupt request (IRG) to the MCU 230 to disable all high-side power switches and dissipate the inductor current by connecting both ends temporarily to power ground 144. Both power ground return legs may be instrumented with current sensors and monitored individually to provide for redundancy as well as status for the current decay in the compromised winding, or leakage currents that may be present while the winding is inserted mid-stream (i.e. current flowing through the bi-directional upstream and downstream switches 280, 290) .

[0049] FIGS. 5A-C represent three possible

configurations to serve DC only applications (FIG. 5A) , single phase AC and DC applications (FIG. 5B), or three- phase AC applications (FIG. 5C). For simplicity, FIG. 5A uses high-power Schottky blocking diodes 158, which may suite higher voltage applications. For low voltage (i.e. below 50V), a back-to-back MOSFET configuration may be more suitable, where the source for each MOSFET is connected together. Alternatively, newer GaN power switches, most commonly connected in cascade and drain-in-common

configurations to provide bi-directional isolation, are suitable to provide both, high-voltage and high-frequency operations. Connections 271, 273, 275, 277, 282, 283, 285, 287, 291, 293, 295, 297 provide access to control the power switches in the matrix 210.

[0050] The H-bridge 270 is indicated in FIG. 4 as connected between upstream and downstream switches 280, 290, and connected to the inductor 400 via connections 146, 148. In FIG. 5A, the H-Bridge 270 includes the power switches 272, 274, 276, 278 and can commonly be used in energy conversion machines 100 to selectively connect ends of the inductor 400 to the positive and negative power connections 142, 144, which can connect to the positive and negative power rails 130, 131, respectfully. The H-Bridge 270 can connect one end 146 of the inductor 400 to either the positive or negative power connections 142, 144, and connect the other end 148 of the inductor 400 to the other one of the positive or negative power connections 142, 144 via selectively turning on and off the power switches 272, 274, 276, 278.

[0051] However, this somewhat conventional H-Bridge 270 does not provide options to connect either one of the ends 146, 148 of the inductor 400 to upstream or downstream inductors 402, 404, 406. The upstream and downstream switches 280, 290 shown in FIG. 4 can be used to provide these connections. The upstream switch 280 can include the power switches 282, 284, 286, 288 (such as power MOSFETs ) , and the downstream switch can include the power switches 292, 294, 296, 298. The H-Bridge 270, the upstream switch 280, and downstream switch 290 may form the power switch matrix 210. The DSP 300 can control the power switches in the power switch matrix 210 to control the path of current through the matrix 210 and thereby, through the IPM and the inductor 400. The power switch matrix 210 is included in the IPM 200, and it should be understood that similar power matrices can be included in the IPMs 202, 204, 206.

[0052] Via DSP 300 control, the IPMs 200, 202, 204, 206 can be configured to connect the four inductors 400, 402, 404, 406 in many configurations, such as those shown in FIGS. 6A-F. It should be understood, that a greater or fewer number of IPMs can be included in the machine 100 as needed to convert between electrical and mechanical energy. In the example shown in FIG. 5A, the power switch matrix 210 has been configured by the DSP 300 to bring current in through the downstream connection 154, through the power switch 298, through the inductor 400 from the end 148 to the end 146, through the power switch 274 and out the negative power supply terminal 144 to the negative power rail 131 (bold lines indicate current path through matrix 210). If it is desired to change the power switch matrix 210

configuration such that the current would enter the

downstream connection 154 and flow out the upstream

connection 152, then the DSP 300 could turn power switch 274 off, and turn power switch 282 on, thereby routing the current through the power switch 282 and out the upstream connection 152. It should be understood that many other configurations can be made by selectively turning off and on the devices in the power switch matrix 210.

[0053] Referring now to FIG. 5B, this configuration of the power switch matrix 210 enables the machine 100 to run on (or generate) either AC or DC, instantaneously, as H- Bridge unidirectional power switches 272, 274, 276, 278 are replaced with bi-directional power switches 272, 274, 276, 278. Additionally, the unidirectional power switches 282, 284, 292, 294, 286, 288, 296, 298 are replaced with bidirectional power switches. However, the power switches 286, 288, 292, 294 are now redundant with power switches 282, 284, 296, 298, since the bi-directional characteristic of the bi-directional power switches allow current to travel through each bi-directional power switch in either direction when they are turned on, and prevents current flow through the power switch when they are turned off. It should be noted that these power switches 286, 288, 292, 294 can remain in the matrix 210, but it is preferred to remove them to minimize complexity of the matrix 210. This

configuration without power switches 286, 288, 292, 294 is shown in FIG. 5B. In this example, the power switch matrix 210 has been configured by the DSP 300 to bring current in through the downstream connection 154, through the power switch 296, through the inductor 400 from the end 148 to the end 146, through the power switch 282 and out the upstream connection 152 (bold lines indicate current path through matrix 210). Selectively controlling the power switches can direct the current through the IPMs in various configurable paths .

[0054] Referring now to FIG. 5C for 3-phase

applications, there are three bi-directional power switches 150 shown for connection of any phase 162, 164, 166 to the inductor end 146, and three bi-directional power switches 150 shown for connection of any phase 162, 164, 166 to the inductor end 148. The power switches 282, 284, 296, 298 are similarly configured as in FIGS. 5A-5B to provide connection of the upstream and downstream connections 152, 154 to the ends 146, 148. This configuration allows the DSP 300 to route current (via switch matrix control) from any one of the phases 162, 164, 166, through the inductor 400 to any one of the phases 162, 164, 166. The inductors 400 can also, like in FIGS. 5A-5B, be connected to upstream and/or downstream inductors 402, 404, 406. Integrating upstream and downstream switches into the architecture enables conventional three-phase applications to take advantage of serialization of inductors and thereby substantially

expanding their operating range, and allows utilization of the full sinusoidal waveform, as it transitions from

positive peak voltage to negative peak voltage, without changing the magnetic polarity in the inductor. [0055] Referring now to FIG. 5D, a table is provided that contains various configurations for a 4-inductor machine 100. The binary values are given in columns 3-0 and represent a 4-bit code transferred to the MCU's 230. The MCU interprets the 4-bit command and can configure the power switch matrix 210 to a pre-determined configuration

associated with the 4-bit code. Some of these configurations are representatively illustrated in FIGS. 6A-6F. It should be understood that the machine 100 can have a greater or fewer number of IPMs as those shown in FIGS. 6A-6F. A practical range limit for the number of IPMs in a machine 100 is at least two, with each IPM having an associated inductor. Three IPMs (and associated inductors) can be used for lower cost three-phase systems as shown in FIG. 7A, where five IPMs can allow a rotation of low inductor count operation as shown in FIG. 7B. Additionally, large machines 100 can be built with as many as 63 IPMs (and associated inductors), and multiple rotors. A possible multiple-rotor configuration is shown in FIGS. 7C-7D with two planes of IPMs with 5 IPMs in each plane.

[0056] FIG. 5D shows 10 different configurations that can be associated with the 10 different 4-bit command codes. The following is a brief description of the potential configurations and how they impact the power switch matrix 201. The BRAKE configuration can be used to brake the rotation of the rotor of the machine 100. The BRAKE command can be executed by turning ON both power switches 274, 278 thereby shorting both ends of the inductor 400 to ground and allowing fast dissipation of the magnetic field surrounding the inductor 400. The OFF command can be executed by turning OFF all power switches thereby preventing flow of current through the IPM. The 4S command can be executed by commanding the IPMs 200, 202, 204, 206 to connect all inductors 400, 402, 404, 406 in series with one end of the series connected to a positive power rail and the other end of the series connected to a negative power rail (see FIG. 6A) . The 3S, 2S, IS commands can be executed by commanding the IPMs 200, 202, 204, 206 to connect 3, 2, or 1 of the inductors 400, 402, 404, 406 in series with each other (see FIG. 6B for an example of the 3S configuration and FIG. 6C for an example of two sets of the 2S configuration), where the remaining inductors are not connected (i.e. connected to an open circuit connection).

[0057] The 2S/2P command can be executed by commanding the IPMs 200, 202, 204, 206 to connect two sets of inductors in series and connecting the two sets in parallel with each other (see FIG. 6C). The 4P, 3P, 2P commands can be

executed by commanding the IPMs 200, 202, 204, 206 to connect 4, 3, 2, or 1 of the inductors 400, 402, 404, 406 in parallel with each other (see FIG. 6F for 4P configuration), where the remaining inductors are not connected (i.e.

connected to an open circuit connection).

[0058] Each of these configurations (4S, 3S, 2S, IS,

2S/2P, 3P, 4P) can be configured by the DSP 300 (via the MCUs 230) through control of the power switch matrices 210. The DSP 300 can cause various configurations to be

configured in a time sequence, thereby systematically and/or periodically cycling through a set of these configurations as needed to maintain optimal operation of the machine 100. FIGS. 6D and 6E illustrate a couple of simple time-sequence operations. FIG. 6D shows first and second series strings with two inductors each (inductors 400, 402 and inductors 404, 406), where, at time "tl", current can be directed to flow through the first series string of inductors 400, 402, with inductors 404, 406 disconnected. At time "t2", current can be directed to flow through the second series string of inductors 404, 406, with inductors 400, 402 disconnected. At time "t3" (not shown) current can again be directed to flow through the first series string of inductors 400, 402, with inductors 404, 406 disconnected. This can be used, for example, to reduce heat generated by the inductors and switches, or for other desired purposes.

[0059] FIG. 6E shows a 4P configuration with four inductors configured for parallel connection with each other. However, not all of the inductors 400, 402, 404, 406 are connected at the same time. As time progresses, another one or more of the parallel configured inductors can be connected with others disconnected. The DSP 300 can cause these time sequenced events to occur at the appropriate times. If only one inductor is connected at a time, then, for example at time "tl" the inductor 400 can be connected between the positive and negative power rails, with

inductors 402, 404, 406 disconnected. At time "t2", inductor 402 can be connected between the positive and negative power rails, with inductors 400, 404, 406

disconnected. At time "t3", inductor 404 can be connected between the positive and negative power rails, with

inductors 400, 402, 406 disconnected. At time "t4", inductor 406 can be connected between the positive and negative power rails, with inductors 400, 402, 404

disconnected. In this example, at time "t5" (not shown), then the sequence could start over by again connecting inductor 400 between the positive and negative power rails, with inductors 402, 404, 406 disconnected. These are merely some very simple examples of the possible configurations as well as time-sequenced configurations that can be achieved with the principles of the current disclosure. It should be understood that time-sequenced configurations for larger machines 100 can get to be very complex to operate the power switch matrix 210 of each of the IPMs in these larger machines 100.

[0060] The machine 100 which embodies principles of the current disclosure can have a universal power connection. The DSP 300 can determine, based on sensor data, what type of power connections is made to the machine and can respond to the sensed data by configuring the inductors 400, 402, 404, 406 to accommodate the type of power detected. For generators, the type of power to be generated can be

selected by the Human Machine Interface. In the case of AC power rails, as stated above, after zero-crossing the power switch matrix can reconnect a positive inductor input from a positive power rail to the negative power rail, from an upstream inductor connection to a downstream inductor connection, from a power rail connection to and upstream or downstream inductor connection, etc., therefore allowing proper current flow and direction to be maintained with minimal losses and minimal noise created by the

reconnections . Reconfiguration of the power switch matrices can be done by the DSP 300 in response to sensed data received from the feedback sensors 310, with these

reconfigurations possibly including connections of upstream and downstream inductors to achieve the optimum machine 100 configuration .

[0061] It will now be fully appreciated that the above disclosure provides several advancements to the art of electric machines. It is to be understood that the various examples described above may be utilized in various

orientations, such as inclined, inverted, horizontal, vertical, etc., and in various configurations, without departing from the principles of the present disclosure. The embodiments illustrated in the drawings are depicted and described merely as examples of useful applications of the principles of the disclosure, which are not limited to any specific details of these embodiments.

[0062] In the above description of the representative examples of the disclosure, directional terms, such as "above," "below," "upper," "lower," etc., are used for convenience in referring to the accompanying drawings.

[0063] Of course, a person skilled in the art would, upon a careful consideration of the above description of representative embodiments, readily appreciate that many modifications, additions, substitutions, deletions, and other changes may be made to these specific embodiments, and such changes are within the scope of the principles of the present disclosure. Accordingly, the foregoing detailed description is to be clearly understood as being given by way of illustration and example only.