Zero to low speed operation of a sensorless brushless DC motor
11251734 · 2022-02-15
Inventors
Cpc classification
H02P23/14
ELECTRICITY
H02P27/04
ELECTRICITY
International classification
H02P23/14
ELECTRICITY
Abstract
A method of operating a Brushless Direct Current Motor (BLDCM), the BLDCM of the type including: a series of concentric independently activated electromagnetic phase coils interacting with a series of permanent magnets to provide relative movement therebetween, the phase coils having temporal periods of activation time and deactivation time, the method including the steps of: (a) activating at least one of the phase coils for a short period of activation; and (b) measuring the voltage response across the phase coil of the deactivated phase coil during the short period of activation to determine the rotor position.
Claims
1. A method of operating a Brushless Direct Current Motor (BLDCM), the BLDCM of the type including: a series of concentric independently activated electromagnetic phase coils interacting with a series of permanent magnets to provide relative movement therebetween, the phase coils having temporal periods of activation time and deactivation time, the method including the step of utilizing a pulse width modulation (PWM) scheme to each of the phase coils using three switching intervals, the switching intervals including: (a) a first on-time switching interval (T1) connecting active phase coils to the voltage supply via a high-side power transistor and to ground via a low-side power transistor; (b) a first off-time switching interval (T2) having a slow phase current fall time, where both of the active phase coils are connected to ground via the low-side power transistors; and (c) a second off-time switching interval (T3) connecting the active phase coils with a reversed polarity having a fast phase current fall time, where the low-side power transistor of the active phase coil is connected to ground and the inactive high-side phase coil is connected to the power supply via a freewheeling diode of the high-side power transistor.
2. A method as claimed in claim 1 further comprising, for at least one phase coil, inserting on-time and off-time open phase winding and power supply voltage measurement pulses at the end of the three switching intervals to determine rotor position (commutation point functions) and rotation direction.
3. A method as claimed in claim 2 further comprising measuring both on-time and off-time open phase winding voltages to deduce commutation points and rotation direction for the BLDCM.
4. A method as claimed in claim 2 further comprising measuring on-time and off-time supply voltages to correct open phase voltage measurements for power supply voltage fluctuations.
5. A method as claimed in claim 4 further comprising correlating the on-time and off-time open phase voltages using on-time and off-time supply voltage measurements.
6. A method as claimed in claim 2 further comprising calculating a series of zero to low speed sensorless BLDCM commutation point functions (CPF) and commutation points using the correlated voltage measurements.
7. A method as claimed in claim 6 further comprising calculating zero to low speed sensorless BLDCM commutation points using an absolute rotor position calculation derived from the correlated voltage measurements.
8. A method as claimed in claim 2 further comprising determining a maximum value of the commutation point function (CPFmax) during operation.
9. A method as claimed in claim 2 further comprising calculating zero to low speed sensorless BLDCM commutation points with on-time (Von) and off-time (Voff) open phase voltage measurements from consecutive PWM cycles.
10. A method as claimed in claim 2 further comprising determining the BLDCM motor rotation direction during zero to low speed sensorless BLDCM operation using the rotor position functions.
11. A method as claimed in claim 1 further comprising activating the low-side power switches of the phase coil during the off-time interval of the pulse measurements to deduce BLDCM commutation points and rotation direction.
12. A method as claimed in claim 1 further comprising utilizing a quasi-field-oriented control (QFOC) algorithm to increase motor torque during zero to low sensorless BLDCM operation.
13. A method of operating a Brushless Direct Current Motor (BLDCM), the BLDCM of the type including: a series of concentric independently activated electromagnetic phase coils interacting with a series of permanent magnets to provide relative movement therebetween, the phase coils having temporal periods of activation time and deactivation time, the method including the steps of: (a) utilizing a pulse width modulation (PWM) scheme to drive each of the phase coils; (b) deriving a series of commutation point functions for each phase coil; (c) determining a maximum value of the commutation point function (CPFmax) during operation; and (d) utilizing the CPFmax value to determine an indication of the BLDCM motor temperature.
14. A method of operating a Brushless Direct Current Motor (BLDCM), the BLDCM of the type including: a series of concentric independently activated electromagnetic phase coils interacting with a series of permanent magnets to provide relative movement therebetween, the phase coils having temporal periods of activation time and deactivation time, the method including the steps of: (a) utilizing a pulse width modulation (PWM) scheme to drive each of the phase coils; (b) deriving a series of commutation point functions for each phase coil; (c) determining a maximum value of the commutation point function (CPFmax) during operation, and (d) utilizing the CPFmax value to determine an indication of the rotor magnetic field strength.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
(1) Embodiments of the invention will now be described, by way of example only, with reference to the accompanying drawings in which:
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)
(18)
(19)
(20)
(21)
(22)
(23)
(24)
(25)
(26)
(27)
(28)
(29)
(30)
(31)
(32)
(33)
(34)
(35)
(36)
(37)
(38)
(39)
(40)
(41)
(42)
(43)
(44)
(45)
(46)
(47)
(48)
(49)
(50)
(51)
(52)
(53)
(54)
(55)
(56)
(57)
(58)
(59)
(60)
(61)
(62)
(63)
(64)
(65)
DETAILED DESCRIPTION
(66) The embodiments are directed to the area of control of electric brushless DC motors (BLDCMs), with a particular focus on the sensorless BLDCM control application, constraint and optimization for compact and efficient dynamically controlled motor systems—one example being cordless electric power tools. The application and technology relate generally to the challenges of creating effective and efficient sensorless BLDCM control to manage the performance and life cycle of such electric motors. A set of challenges for sensorless BLDCM controllers are apparent, which translate across a large set of applications and realizations, these are:
(67) Motor start at standstill and in motion; Motor start with reversed rotor momentum; Motor start in presence of external load disturbances, gearbox backlash; Motor start time and response; Motor operation and torque control down to stall (0 RPM), at low speed, with reversed rotor momentum, dynamic loads and external disturbances; Controller operation with a broader range of BLDCMs available on the market (salient and non-salient pole, wye and delta winding, asymmetric stator phase winding inductances and resistances); Controller efficiency during motor operation, regenerative motor braking; Energy return to the power source during regenerative motor braking; Controller sensitivity to supply voltage and temperature variations, noise; Controller setup and tuning, integration time; Controller electronic circuit design complexity, size, cost; Motor and controller system reliability, operation and maintenance cost; Rotor magnetic field strength measurement in real time and monitoring over life time of operation; Motor temperature measurement and monitoring in real time; Printed circuit board (PCB) layout optimization in three-phase power control applications.
(68) The embodiments provide for sensorless BLDCM controller designs that have improved performance, operational behaviour and greater possibility across one or more of these areas of challenge. As such, the disclosure of the embodiments is applicable to a wide area, and more broadly applicable to the application of BLDCMs across applications, including and not limited to power tools, locomotion, mobility, robotics, automation and control, automotive, medical, consumer, computer, hobby, etc. A general primer of the breadth and applicability of the areas of interest and application of BLDCMs can be found in The following prior art references provide general background information in the field.
(69) [1], [2], [3].
(70) The embodiments provide an initial rotor position detection which presents a method that is insensitive to phase winding asymmetries in BLDCMs.
(71) The embodiments provide an improved inductance variation technique employing only voltage measurements to detect the initial rotor position and as such do not require any additional electronic circuitry. The disclosed technique also makes it possible to detect the initial rotor position of a wide range of BLDCMs extremely fast, typically in <1 ms, whilst doing so with greater accuracy and precision.
(72) The embodiments specifically address the challenges of detecting the initial rotor position with the motor already spinning at very low speeds, for example, below the BEMF voltage measurement threshold, or when starting with a reversed rotor momentum. These sensorless brushless motor control challenges are frequently encountered in applications, such as mobility equipment and cordless electric power tools.
(73) The synchronous rectification method disclosed in the embodiments provides a near optimal, software controlled approach that integrates seamlessly into the sensorless BLDCM operation during PWM switching and dynamic regenerative motor braking. This is achieved without any additional electronic hardware circuitry, leading to further improvements in controller output efficiency under all motor loads. It is evident that the disclosed synchronous rectification method can also be applied to any other inductive PWM switching applications. For example, DC/DC power supplies, brushed and sensored brushless motors.
(74) The novel software controlled synchronous rectification technique disclosed in the embodiments can be used to increase controller efficiency and the amount of energy recycled back into the power source during motor braking. It can thus improve the performance of battery operated motor applications, such as mobility equipment, by increasing the time before battery recharging is required.
(75) An innovative step of the disclosed controller embodiments is the ability to measure a BLDCM's rotor magnetic field strength during operation without requiring any additional hardware or electronic circuitry. It is also possible to monitor the rotor magnetic field strength over the entire life cycle of operation. In many brushless motor applications such a built-in, automatic rotor measurement can have the advantage of decreasing operation and maintenance costs and increasing system performance.
(76) The embodiments disclose a novel BLDC motor temperature measurement that does not require any additional temperature sensing equipment, hardware or electronic circuitry. The disclosed method uses electrical motor measurements obtained during sensorless BLDCM mode of operation from which reliable motor temperature measurements in real time of operation are obtained.
(77) To address power and thermal loss challenges in three-phase power control applications, the embodiments also disclose a novel PCB layout design consisting of power transistors placed in a radially symmetrical configuration that achieves a significant reduction in the PCB copper track resistance in each phase compared to the prior art, reducing power and thermal losses and increasing controller efficiency.
(78) The sensorless brushless DC motor controller of the preferred embodiment exploits the inductance variations which exist in BLDC motors due to rotor inherent and magnetic saturation (RIMS) saliency. The disclosed controller methods improve the accuracy, precision and speed of initial rotor position detection compared to the prior art and extend the inductance variation properties to novel commutation point detection methods for sensorless operation of BLDC motors at zero and high speeds. The disclosed sensorless BLDCM control operation is highly robust during start-up and operation at near zero speed, providing torque control down to motor stall and operation with reversed rotor momentum with a broader range of wye and delta configuration BLDCMs. The disclosed sensorless BLDCM controller is also independent of motor parameters and uses simplified hardware circuitry that requires only voltage measurements, making it easier and cheaper to implement in practical applications.
(79) As will be further described below, the embodiments disclose a hybrid sensorless BLDCMs control approach, as illustrated by the flow chart 1 of
(80) To address the controller challenge of robust sensorless brushless operation with the broadest possible range of BLDC motors, two different controller embodiments are disclosed with the second embodiment denoted with the Δ suffix.
(81) To proceed with the description, it is instructive to introduce some brushless motor controller background information.
(82) A Sensorless BLDCM Controller Circuit
(83) The basic equivalent circuit for the control of the BLDCM according to the disclosed embodiments is illustrated 20 in
(84) BLDC motor 21 with a stator consisting of plurality of coil phase windings, wye or delta phase winding configuration (
(85) BLDC motor switching control circuit 23 consisting of plurality of semiconductor power transistor switches (A_L, A_H, B_L, B_H, C_L, C_H) such as MOSFETs or IGBTs to control the phase winding currents in a synchronized manner with the rotor position and includes freewheeling semiconductor diodes e.g. 24 to conduct off-time switching inductive currents.
(86) Power transistor gate control circuit consisting of plurality of gate drivers (A_L, A_H, B_L, B_H, C_L, C_H) 27 to optimally control the power transistors 23 during switching operation. Additionally, an Analog to digital converter circuit 29, consisting of high speed plurality of analog voltage measurements (Va, Vb, Vc), which can include voltage resistor divider networks to reduce the sampled analog voltages to an acceptable level for measurement with an ADC and capacitors to reduce measured analog voltage bandwidth and filter out electrical noise.
(87) A Microcontroller 28 provides various functionalities according to the various embodiments. The controller may comprise an integrated high speed ADC circuit, volatile memory such as DRAM, and non-volatile memory such as PROM, EPROM, EEPROM, FLASH, MRAM, PCRAM, and other functionalities such as input and output interfacing, and communication ports etc.
(88) Electrical voltage source (Vs) such as a transformer or a switching power supply or a battery.
(89) Rotor Inherent Saliency
(90) If the flux from the permanent magnets in the rotor is large enough to magnetically saturate the stator iron in a BLDC motor, then a variation in the rotor's direct and quadrature magnetic axes with respect to the stator coils leads to a variation in phase inductance shown 30 in
(91) Magnetic Saturation Saliency
(92) If current is flowing through the phase coils of a permanent magnet BLDC motor, then the flux produced by the phase coils in the stator interacts with the flux from the permanent magnet rotor, leading to additional variations in the troughs of the phase inductances as illustrated 40 in
(93) Inductance Variation Measurement
(94) In the embodiments, phase coils inductance variation due to RIMS saliency is measured indirectly as a voltage variation at the output of inactive (open) phase coil after applying a short duration current pulse to a pair of phase coils. This open phase winding voltage measurement technique is different to the prior art, such as disclosed in U.S. Pat. Nos. 5,028,852, 5,191,270, 6,172,498, 6,850,863, 7,023,155, 7,489,096, 7,592,761, 7,944,159, WO 2012/010065, U.S. Pat. Nos. 8,400,086, 8,796,974, where current variations due to phase inductance changes are measured in deducing the rotor position, for example as illustrated in
(95) Some examples of inductance variation based measurement techniques from the prior art employing open phase voltage measurements are disclosed in U.S. Pat. Nos. 6,344,721, 6,650,082 and US2014/0015458. It is evident that these techniques only employ voltage measurements at the open phase winding during the on-time period of pulse to estimate the rotor position, as shown in FIG. 7 of disclosure US2014/0015458.
(96) In contrast to the prior art, the present embodiments use both the on-time and off-time open phase winding measurements in the generated pulses to ascertain the rotor position. In an attempt to address a set of challenges in the area of sensorless BLDCM controllers, the embodiments offer several advantages, including: increased rotor position angle detection accuracy; faster rotor position detection time; rotor position detection independent of rotor speed; insensitivity to stator phase winding inductance and resistance asymmetries in BLDC motors.
(97) Initial Rotor Position Detection
(98) This section discloses the initial rotor position detection methods developed. A description of the pulse generation and on-time and off-time phase voltage measurements for each phase coil pair combination is presented first. This is followed by a description of novel rotor position functions (RPF) and rotation direction functions (RDF) used to calculate accurate rotor position and spin direction.
(99) The key constituents of the initial rotor position detection methods disclosed in this section are: On-time and off-time open phase voltage measurement; Rotor position functions (RPF); Enhanced rotor position functions (ERPF); Absolute angle rotor position calculation (θ); and Rotation direction functions (RDF).
(100) The operation flowcharts of the three different initial rotation direction detection methods disclosed in this section are illustrated in
(101) (a) On-Time and Off-Time Open Phase Voltage Measurement
(102) Both the on-time and the off-time voltage responses are measured at the inactive phase coil winding for each of the six possible phase coil pair combination (B+/C−, C+/B−, C+/A−, A+/C−, A+/B−, B+/A−) and then consequently used to calculate the rotor position. This is one of the key advantages. An example of the voltage and current waveforms generated in the disclosed method for the case when B+/C− phase coils (B_H and C_L power transistors in circuit diagram in
(103) (i) On-Time Measurement Pulse Interval
(104) The on-time (Ton) pulse measurement duration, shown in
(105) During the on-time interval of the generated pulse, the Von measurement is taken at the open phase winding. In the example shown in
(106) In addition to the open phase voltage measurement (Von), the supply voltage rail (Vs) is also measured during this on-time pulse interval. This is performed to correct the open phase voltage measurements of all phase coil pair combinations for any fluctuations in the supply voltages, for example as shown in
(107) (ii) Off-Time Measurement Pulse Interval
(108) After the on-time interval has elapsed the B_H and C_L power transistor devices are turned off and B_L transistor is turned on, as illustrated in the example of
(109) The off-time (Voff) voltage measurement is performed as soon as the turn-off switching transients in the open phase winding have decayed away. In practice, this was found to typically occur around 5 μsec-15 μsec after the end of on-time (Ton) period. After the completion of Voff open phase winding voltage measurement, the bottom power transistor (B_L) is turned-off, as illustrated in
(110) During the off-time interval the current i(t) in phase coils B and C flows through the freewheeling diode of power transistor C_H until it decays to zero, as depicted in
(111) To correlate the measured off-time open phase voltage to the one measured during on-time pulse interval, the embodiment also performs a voltage measurement at the C phase winding, as shown in
(112) (iii) On-time and Off-time Measurement Pulse Interval Extensions
(113) It should be noted that it is also possible to use the same method to measure the on-time (Von) and off-time (Voff) open phase voltages with respect to the supply voltage (Vs) rail, instead of the ground (0V) voltage rail. In this case, during the off-time pulse measurement interval, the phase coil with the positive voltage polarity is actively connected to the supply voltage (Vs) rail via the high-side power transistor, by turning it on. The phase coil with the negative voltage polarity is disconnected from the ground rail (0V) by turning-off the low-side power transistor. This allows the decaying phase current i(t) to circulate to the ground voltage rail via the low-side power transistor's internal freewheeling diode. To illustrate this variation in the disclosed off-time interval pulse measurement method with a practical example in
(114) As a further extension to the disclosed on-time (Von) and off-time (Voff) open phase voltage measurements, it is also possible to make the off-time (Voff) open phase voltage measurements during the off-time pulse interval with all power transistors turned-off. This variation in the disclosed method during the off-time pulse measurement interval, illustrated in
(115) (b) On-Time and Off-Time Voltage Measurement Sequence
(116) To determine the initial rotor position, a sequence of on-time and off-time open phase winding and supply voltage measurements corresponding to the six possible pairs of phase combinations is performed first. An example sequence captured from a practical BLDC motor is shown in
(117) Since the pulse on-time durations are very short, typically in the range between 20 μsec-70 μsec, no net torque is imposed on the BLDCM rotor to cause it to turn and thus the rotor remains stationary. The entire measurement process takes <1 ms to complete (typ. ˜600 us) which is very short.
(118) (i) On-Time and Off-Time Open Phase Voltage Measurements
(119) Table 1 below shows the six pairs of on-time and off-time open phase winding voltage measurements, obtained from the measurement sequence shown in
(120) TABLE-US-00001 TABLE 1 On-time and Off-time open phase voltage measurements On-time Off-time Active Phase Open Phase Open Phase Coil Pair Voltage Voltage B+/C− VonAbc VoffAbc C+/B− VonAcb VoffAcb C+/A− VonBca VoffBca A+/C− VonBac VoffBac A+/B− VonCab VoffCab B+/A− VonCba VoffCba
(121) For example, the notation VonAbc refers to open phase winding A on-time voltage measurement with phase coil B connected to Vs via B_H switch and phase coil C connected to the ground via C_L switch. Similarly, the notation VoffAbc refers to open phase winding A off-time voltage measurement with phase coil C freewheeling to Vs+Vd via C_H switch diode and phase coil B connected to the ground via B_L switch.
(122) (ii) On-Time and Off-Time Supply Voltage Measurements
(123) The six pairs of supply voltage measurements during the on-time and off-time pulse intervals shown in Table 2 are also performed during the measurement sequence presented in
(124) TABLE-US-00002 TABLE 2 On-time and Off-time open phase voltage measurements On-time Off-time Active Phase Phase Supply Phase Supply Coil Pair Voltage Voltage B+/C− Vs_BC Vsoff_BC C+/B− Vs_CB Vsoff_CB C+/A− Vs_CA Vsoff_CA A+/C− Vs_AC Vsoff_AC A+/B− Vs_AB Vsoff_AB B+/A− Vs_BA Vsoff_BA
(125) Where, for example, the notation Vs_BC refers to on-time supply voltage (Vs) measurement with phase coil B connected to Vs via B_H switch and phase coil C connected to the ground via C_L switch. Similarly, the notation Vsoff BC refers to off-time supply voltage measurement (Vs+Vd) with phase coil C freewheeling to Vs+Vd voltage level via C_H switch diode and phase coil B connected to the ground via B_L switch.
(126) (iii) On-Time and Off-Time Voltage Measurement Correlation
(127) In the disclosed method, the measured on-time and off-time open phase voltages are correlated to each other for the following key reasons: 1) To scale down the off-time (Voff) open phase winding voltage measurements that are referenced to voltage level Vs+Vd, instead of Vs as in on-time (Von) open phase winding measurements; 2) To compensate on-time (Von) and off-time (Voff) open phase winding voltage measurements for any fluctuations in the power supply, as shown in
(128) Correlation of Voff to Von measurements is performed via the scaling calculation Vs/Vsoff using their respective Vs and Vsoff supply voltage measurements, as shown in the following equations for each phase coil pair:
(129) TABLE-US-00003 VoffAbc′ = VoffAbc .Math. Vs_BC/Vsoff BC (1) VoffAcb′ = VoffAcb .Math. Vs_BC/Vsoff CB (2) VoffBca′ = VoffBca .Math. Vs_BC/Vsoff CA (3) VoffBac′ = VoffBac .Math. Vs_BC/Vsoff AC (4) VoffCab′ = VoffCab .Math. Vs_BC/Vsoff AB (5) VoffCba′ = VoffCba .Math. Vs_BC/Vsoff BA (6)
(130) It should be noted that, in this embodiment, all Voff open phase measurements in the measurement sequence are correlated to the open phase measurement of the first phase coil pair. In the example measurement sequence shown in
(131) To prevent fluctuations in supply voltage rail (Vs) from affecting the Von and Voff open phase winding measurements in the entire measurement sequence, as shown in
(132) TABLE-US-00004 VonAbc′ = VonAbc (7) VonAcb′ = VonAcb .Math. Vs_BC/Vs_CB (8) VonBca′ = VonBca .Math. Vs_BC/Vs_CA (9) VonBac′ = VonBac .Math. Vs_BC/Vs_AC (10) VonCab′ = VonCab .Math. Vs_BC/Vs_AB (11) VonCba′ = VonCba .Math. Vs_BC/Vs_BA (12)
(133) The open phase voltage correlation step is shown in initial rotor position detection operation flowcharts in
(134) (iv) Stator Core Demagnetization
(135) During testing of the disclosed sensorless BLDCM controller, it was found that high motor currents can leave a net residual magnetic flux in the iron stator core in BLDCM motors. This can lead to reduced accuracy and precision of the detected initial rotor positions. To address this issue, a novel and improved measurement technique was developed that consists of the application of very short duration pulses before the main on-time and off-time measurement pulses, as shown in
(136) Pulse durations in the range between 5 μsec-15 μsec were tested to work well in practice. The purpose of these shorter preceding pulses is to quickly demagnetize the iron stator core in the event that it contains some net residual magnetic flux fields from a previous sensorless BLDCM operation. The accuracy and precision of Von and Voff measurements and thus the resultant rotor position detection is improved in practical application.
(137) In time critical initial rotor position detection applications, where the fastest possible detection time is required, it was found that equal improvements were also obtained by applying the shorter duration iron stator core demagnetization pulses to only every second phase coil pair (B+/C−, C+/A−, A+/B−) as shown in
(138) The disclosed short preceding pulses, which offer a performance improvement, are not critical to the working of the overall embodiment. Thus they can also be completely omitted from the on-time and off-time phase voltage measurements pulse sequence.
(139) (c) Rotor Position Functions
(140) The six pairs of correlated Von and Voff open phase winding measurements obtained from the sequence of voltage pulses, as shown in application examples in
(141) 1) Increased rotor position angle detection accuracy to within ±30° sectors, compared to ±60° sectors commonly disclosed in the prior art methods, thus increasing motor start-up performance;
(142) 2) Absolute rotor position calculation to with few angular degrees using additional trigonometric calculations;
(143) 3) Rotor position detection independent of rotor speed, making detection possible at standstill or in motion;
(144) 4) Insensitivity to stator phase winding inductance and resistance asymmetries in BLDC motors
(145) Two embodiments used to calculate the RPFs are disclosed, with the 2nd embodiment denoted with the Δ suffix.
(146) (i) Rotor Position Function Calculations (1st Embodiment)
(147) According to the first embodiment, the rotor position functions (PA, PB, PC) consisting of correlated Von and Voff open phase winding measurements are calculated with:
PA=(VonCab−VoffCab)+(VonCba−VoffCba)+(VonBca−VoffBca)+(VonBac−VoffBac)−2.Math.(VonAbc−VoffAbc)−2.Math.(VonAcb−VoffAcb) (13)
PB=(VonAbc−VoffAbc)+(VonAcb−VoffAcb)+(VonCab−VoffCab)+(VonCba−VoffCba)−2.Math.(VonBca−VoffBca)−2.Math.(VonBac−VoffBac) (14)
PC=(VonBca−VoffBca)+(VonBac−VoffBac)+(VonAbc−VoffAbc)+(VonAcb−VoffAcb)−2.Math.(VonCab−VoffCab)−2.Math.(VonCba−VoffCba) (15)
(148) An application example of the output rotor positions calculated by these functions in a BLDC motor is presented in
(149) In addition, a set of rotor position functions (PA_30, PB_30, PC_30) shifted by +30° compared to (PA, PB, PC) are also disclosed. They are used to further improve the detected rotor position accuracy to within ±30° sectors and are calculated with the following functions consisting of correlated Von and Voff measurements:
PA_30=(VonCab−VoffCab)+(VonCba−VoffCba)−(VonAbc−VoffAbc)−(VonAcb−VoffAcb) (16)
PB_30=(VonAbc−VoffAbc)+(VonAcb−VoffAcb)−(VonBca−VoffBca)−(VonBac−VoffBac) (17)
PC_30=(VonBca−VoffBca)+(VonBac−VoffBac)−(VonCab−VoffCab)−(VonCba−VoffCba) (18)
(150) An example of the +30° shifted output rotor positions calculated by these functions in a practical BLDC motor application is shown in
(151) (ii) −30° Shifted Rotor Position Function Calculations (2nd Embodiment (Δ))
(152) In certain challenge applications of initial rotor position detection of BLDC motors, it was found advantageous to employ rotor positions functions shifted by −30° compared to the functions (PA, PB, PC) and (PA_30, PB_30, PC_30) disclosed in the first embodiment. The reasons for this will become more apparent below. Apart from this −30° phase shift, these rotor position functions have the same properties as the sets of functions (PA, PB, PC) and (PA_30, PB_30, PC_30) disclosed in the first embodiment.
(153) According to the second embodiment of this invention, the −30° shifted rotor position functions (PA_Δ, PB_Δ, PC_Δ) consisting of correlated Von and Voff open phase winding measurements are calculated with:
PA_Δ=(VonBca−VoffBca)+(VonBac−VoffBac)−(VonAbc−VoffAbc)−(VonAcb−VoffAcb) (19)
PB_Δ=(VonCab−VoffCab)+(VonCba−VoffCba)−(VonBca−VoffBca)−(VonBac−VoffBac) (20)
PC_Δ=(VonAbc−VoffAbc)+(VonAcb−VoffAc)−(VonCab−VoffCab)−(VonCba−VoffCba) (21)
(154) Similarly, according to this second embodiment, the set of rotor position functions (PA_30_Δ, PB_30_Δ, PC_30_Δ) shifted by +30° compared to (PA_Δ, PB_Δ, PC_Δ) are calculated with the following functions consisting of correlated Von and Voff measurements:
PA_30_Δ=PA (22)
PB_30_Δ=PB (23)
PC_30_Δ=PC (24)
(155) (iii) Rotor Position Detection Implementation with RPFs
(156) The following section presents the implementation of the disclosed rotor position detection. The RPF calculations from the first 1st embodiment are presented. For the implementation of the 2nd embodiment, the function equations with the Δ suffix are used, for example PA_Δ instead of PA. The steps used to determine the rotor position within ±30° sectors are:
(157) 1) Calculate (PA, PB, PC) RPFs using the six pairs of correlated Von and Voff open phase winding measurements as inputs.
(158) 2) Determine initial rotor position to within 60° sector by checking the sign of the RPFs, as shown in Table 3. The corresponding commutation state number is obtained.
(159) 3) Refine rotor position to within 30° sector with (PA_30, PB_30, PC_30) RPFs corresponding to commutation state determined in step 1. The sign of this RPF determines the ±30° sector within current 60° sector, as shown in Table 4.
(160) TABLE-US-00005 TABLE 3 Initial rotor position estimate within 60° sector using (PA, PB, PC) RPFs Rotor PA RPF PB RPF PC RPF Position Commutation Sign Sign Sign Sector θ State Check Check Check 0-60° 0 PA ≥ 0 PB < 0 PC ≥ 0 60-120° 1 PA ≥ 0 PB < 0 PC < 0 120-180° 2 PA ≥ 0 PB ≥ 0 PC < 0 180-240° 3 PA < 0 PB ≥ 0 PC < 0 240-300° 4 PA < 0 PB ≥ 0 PC ≥ 0 300-360° 5 PA < 0 PB < 0 PC ≥ 0
(161) TABLE-US-00006 TABLE 4 Refined rotor position estimate within 30° sectors using (PA_30, PB_30, PC_30) RPFs Rotor RPF Position Commutation Sign Sector θ State Check 0-30° 0 PA_30 < 0 30-60° 0 PA_30 ≥ 0 60-90° 1 PC_30 ≥ 0 90-120° 1 PC_30 < 0 120-150° 2 PB_30 < 0 150-180° 2 PB_30 ≥ 0 180-210° 3 PA_30 ≥ 0 210-240° 3 PA_30 < 0 240-270° 4 PC_30 < 0 270-300° 4 PC_30 ≥ 0 300-330° 5 PB_30 ≥ 0 330-360° 5 PB_30 < 0
(162) This initial rotor position detection method is also presented in the operation flowchart shown in
(163) (d) Enhanced Rotor Position Functions
(164) During development of the disclosed sensorless BLDCM controller, it was found that some BLDC motors exhibit undesirable characteristics which make it more difficult to detect accurate and precise rotor position using the rotor position function calculations disclosed in the 1st embodiment (PA, PB, PC) and (PA_30, PB_30, PC_30), and the 2nd embodiment (PA_Δ, PB_Δ, PC_Δ) and (PA_30_Δ, PB_30_Δ, PC_30_Δ). These non-ideal characteristics can exist in some BLDC motor due to their physical construction, the level of rotor magnetic field strength and effects such as mutual inductance.
(165) Consequently, a set of additional initial rotor position detection embodiments have also been developed to address this challenge. These consist of a set of enhanced rotor position functions (ERPF) developed to work in conjunction with the previously disclosed RPFs that have similar properties.
(166) As with RPFs, two ERPF embodiments are disclosed, with the 2nd embodiment denoted with the A suffix.
(167) (i) Enhanced Rotor Position Function Calculations (1st Embodiment)
(168) The enhanced rotor position functions (EPA0, EPC1, EPB2, EPA3, EPC4, EPB5), consisting of correlated Von and Voff open phase winding measurements are calculated with:
EPA0=(VonCba−VoffCab)+(VonBac−VoffBca)−2.Math.(VonAcb−VoffAbc) (300°≤θ<60°) (25)
EPC1=(VonBca−VoffBac)+(VonAbc−VoffAcb)−2.Math.(VonCab−VoffCba) (0°≤θ<120°) (26)
EPB2=(VonAcb−VoffAbc)+(VonCba−VoffCab)−2.Math.(VonBac−VoffBca) (60°≤θ<180°) (27)
EPA3=(VonCab−VoffCba)+(VonBca−VoffBac)−2.Math.(VonAbc−VoffAcb) (120°≤θ<240°) (28)
EPC4=(VonBac−VoffBca)+(VonAcb−VoffAbc)−2.Math.(VonCba−VoffCab) (180°≤θ<300°) (29)
EPB5=(VonAbc−VoffAcb)+(VonCab−VoffCba)−2.Math.(VonBca−VoffBac) (240°≤θ<360°) (30)
(169) These enhanced rotor position functions are applied to each of the six 60° rotor position sectors to improve the zero crossings of the calculated rotor position functions, as shown in the practical example in
(170) In addition, a set of rotor position functions (EPA0_30, EPC1_30, EPB2_30, EPA3_30, EPC4_30, EPB5_30) shifted by +30° compared to (EPA0, EPC1, EPB2, EPA3, EPC4, EPB5) are also disclosed in this embodiment. As with the RPFs, they are used to further improve the detected rotor position accuracy to within ±30° sectors and are calculated with the following functions consisting of correlated Von and Voff measurements:
EPA0_30=(VonCab−VoffCba)−(VonAcb−VoffAbc) (0°≤θ<60°) (31)
EPC1_30=(VonBac−VoffBca)−(VonCab−VoffCba) (60°≤θ<120°) (32)
EPB2_30=(VonAbc−VoffAcb)−(VonBac−VoffBca) (120°≤θ<180°) (33)
EPA3_30=(VonCba−VoffCab)−(VonAbc−VoffAcb) (180°≤θ<240°) (34)
EPC4_30=(VonBca−VoffBac)−(VonCba−VoffCab) (240°≤θ<300°) (35)
EPB5_30=(VonAcb−VoffAbc)−(VonBca−VoffBac) (300°≤θ<360°) (36)
(171) (ii) −30° Shifted Enhanced Rotor Position Function Calculations (2nd Embodiment (Δ))
(172) The enhanced rotor position functions (EPA0_Δ, EPC1_Δ, EPB2_Δ, EPA3_Δ, EPC4_Δ, EPB5_Δ) shifted by −30° compared to the functions (EPA0, EPC1, EPB2, EPA3, EPC4, EPB5) are calculated with the following functions consisting of correlated Von and Voff measurements:
EPA0_Δ=(VonBca−VoffBac)−(VonAcb−VoffAbc) (300°≤θ<60°) (37)
EPC1_Δ=(VonAcb−VoffAbc)−(VonCab−VoffCba) (0°≤θ<120°) (38)
EPB2_Δ=(VonCab−VoffCba)−(VonBac−VoffBca) (60°≤θ<180°) (39)
EPA3_Δ=(VonBac−VoffBca)−(VonAbc−VoffAcb) (120°≤θ<240°) (40)
EPC4_Δ=(VonAbc−VoffAcb)−(VonCba−VoffCab) (180°≤θ<300°) (41)
EPB5_Δ=(VonCba−VoffCab)−(VonBca−VoffBac) (240°≤θ<360°) (42)
(173) Similarly, a set of rotor position functions (EPA0_Δ_30, EPC1_Δ_30, EPB2_Δ_30, EPA3_Δ_30, EPC4_Δ_30, EPB5_Δ_30) shifted by +30° compared to (EPA0_Δ, EPC1_Δ, EPB2_Δ, EPA3_Δ, EPC4_Δ, EPB5_Δ) are calculated with the following functions consisting of correlated Von and Voff measurements:
EPA0_Δ_30=EPA0 (0°≤θ<60°) (43)
EPC1_Δ_30=EPC1 (60°≤θ<120°) (44)
EPB2_Δ_30=EPB2 (120°≤θ<180°) (45)
EPA3_Δ_30=EPA3 (180°≤θ<240°) (46)
EPC4_Δ_30=EPC4 (240°≤θ<300°) (47)
EPB5_Δ_30=EPB5 (300°≤θ<360°) (48)
(174) (iii) Rotor Position Detection Implementation with RPFs and ERPFs
(175) If improvements in the accuracy of rotor position calculations are required to overcome the flat regions and kinks in RPF zero-crossing that exist in some BLDC motors, for example as shown in
(176) 1) Calculate (PA, PB, PC) using the 6 sets of correlated Von and Voff measurements as inputs.
(177) 2) Determine initial rotor position to within 60° sector by checking the sign of (PA, PB, PC) RPFs, as shown in Table 3. The corresponding commutation state number is obtained.
(178) 3) Improve the accuracy of rotor position estimate obtained in step 2, by checking (EPA0, EPC1, EPB2, EPA3, EPC4, EPB5) ERPFs as outlined in Table 5. A refined output rotor position, 0 and commutation state are obtained after this step.
(179) 4) Refine rotor position to within 30° sector with (EPA0_30, EPC1_30, EPB2_30, EPA3_30, EPC4_30, EPB5_30) ERPF corresponding to the commutation state determined in step 3. The sign of this ERPF determines rotor position to 130° within current 60° sector, as shown in Table 6.
(180) TABLE-US-00007 TABLE 5 Refined rotor position using (EPA0, EPC1, EPB2, EPA3, EPC4, EPB5) ERPFs Input Rotor Pos. Output Rotor Pos. Comm. ERPFs ERPF Comm. State Sector θ Comparison Sign Check State Sector θ 0 0-60° EPA0 < EPC1 EPA0 < 0 5 300-360° EPA0 ≥ 0 0 0-60° 0 0-60° EPA0 ≥ EPC1 EPC1 < 0 1 60-120° EPC1 ≥ 0 0 0-60° 1 60-120° EPC1 > EPB2 EPC1 < 0 1 60-120° EPC1 ≥ 0 0 0-60° 1 60-120° EPC1 ≤ EPB2 EPB2 < 0 1 60-120° EPB2 ≥ 0 2 120-180° 2 120-180° EPB2 < EPA3 EPB2 < 0 1 60-120° EPB2 ≥ 0 2 120-180° 2 120-180° EPB2 ≥ EPA3 EPA3 < 0 3 180-240° EPA3 ≥ 0 2 120-180° 3 180-240° EPA3 > EPC4 EPA3 < 0 3 180-240° EPA3 ≥ 0 2 120-180° 3 180-240° EPA3 ≤ EPC4 EPC4 < 0 3 180-240° EPC4 ≥ 0 4 240-300° 4 240-300° EPC4 < EPB5 EPC4 < 0 3 180-240° EPC4 ≥ 0 4 240-300° 4 240-300° EPC4 ≥ EPB5 EPB5 < 0 5 300-360° EPB5 ≥ 0 4 240-300° 5 300-360° EPB5 > EPA0 EPB5 < 0 5 300-360° EPB5 ≥ 0 4 240-300° 5 300-360° EPB5 ≤ EPA0 EPA0 < 0 5 300-360° EPA0 ≥ 0 0 0-60°
(181) TABLE-US-00008 TABLE 6 Refined rotor position estimate within 30° sectors using (EPA0_30, EPC1_30, EPB2_30, EPA3_30, EPC4_30, EPB5_30) ERPFs Rotor ERPF Position Commutation Sign Sector θ State Check 0-30° 0 EPA0_30 < 0 30-60° 0 EPA0_30 ≥ 0 60-90° 1 EPC1_30 ≥ 0 90-120° 1 EPC1_30 < 0 120-150° 2 EPB2_30 < 0 150-180° 2 EPB2_30 ≥ 0 180-210° 3 EPA3_30 ≥ 0 210-240° 3 EPA3_30 < 0 240-270° 4 EPC4_30 < 0 270-300° 4 EPC4_30 ≥ 0 300-330° 5 EPB5_30 ≥ 0 330-360° 5 EPB5_30 < 0
(182) This initial rotor position detection method employing both RPF and ERPF calculations is presented in the operation flowchart shown in
(183) (iv) Enhanced Rotor Position Functions Extensions
(184) In application with certain types of BLDC motors, such as internal permanent magnet (IPM) motors, it was found that the calculation order of the enhanced rotor position functions (EPA0, EPC1, EPB2, EPA3, EPC4, EPB5), and +30° shifted functions (EPA0_30, EPC1_30, EPB2_30, EPA3_30, EPC4_30, EPB5_30) of the 1st and 2nd (Δ) embodiments had to be modified, in order to obtain the same effect. This is achieved by swapping the enhanced rotor position function calculations with the same phase coil pairs (eg. EPAx, EPBx, EPCx) as outlined below:
EPA0x=EPA3x (0°≤θ<60°)
EPC1x=EPC4x(60°≤θ<120°)
EPB2x=EPB5x(120°≤θ<180°)
EPA3x=EPA0x (180°≤θ<240°)
EPC4x=EPC1x(240°≤θ<300°)
EPB5x=EPB2x(300°≤θ<360°)
(185) Where: x={‘ ’, _30, _Δ, _Δ_30}
(186) Thus, this calculation swap applies to all of the following disclosed enhanced rotor position function calculations, eg. normal (no shift) and shifted by +30°, of the 1st and 2nd (Δ) embodiments: (EPA0, EPC1, EPB2, EPA3, EPC4, EPB5)| (EPA0_30, EPC1_30, EPB2_30, EPA3_30, EPC4_30, EPB5_30) (EPA0_Δ, EPC1_Δ, EPB2_Δ, EPA3_Δ, EPC4_Δ, EPB5_Δ) (EPA0_Δ_30, EPC1_Δ_30, EPB2_Δ_30, EPA3_Δ_30, EPC4_Δ_30, EPB5_Δ_30)
(187) (e) Absolute Rotor Position Calculation
(188) Accurate rotor position detection within ±30° sectors is possible with simple interpretations of the zero-crossing points and intersection points of the rotor position functions, as disclosed in the previous sections. This embodiment presents a method employing trigonometric calculations to improve the angular resolution to within a few degrees of the actual electrical rotor position. The complex vector space is an example of such a method that projects (PA, PB, PC) RPF voltage magnitudes from which the resultant rotor position vector (PR) is calculated with the following equation:
PR=√[(PA−PB/2−PC/2).sup.2+¾.Math.(PB−PC).sup.2].Math.exp{j.Math.tan.sup.−1[√(3/2).Math.(PB−PC)/(PA−PB/2−PC/2)]} (49)
(189) The absolute rotor position (θ) is then given by:
θ=tan.sup.−1[√(3/2).Math.(PB−PC)/(PA−PB/2−PC/2)] (50)
(190) For the implementation of the 2nd embodiment of the disclosed controller, the function equations with the Δ suffix are used, for example PA_Δ instead of PA.
(191) (f) Rotation Direction Functions
(192) When starting with a BLDC motor in motion, the forward or reverse rotor spin direction is determined with a set of developed rotation direction functions (RDF), which are calculated using correlated off-time (Voff) open phase winding measurements. These disclosed rotation direction functions evaluate to either a positive or negative value depending on the rotor spin direction. As with RPFs and ERPFs, two embodiments are disclosed, with the 2nd embodiment denoted with the Δ suffix. It should be noted that it is equally possible to calculate the rotor spin direction with the same equations using correlated on-time (Von) open phase measurements, the only difference being that the signs of the output functions have the opposite value
(193) (i) Rotor Direction Function Calculations (1st Embodiment)
(194) The rotation direction functions (RA, RB, RC) are calculated with the following equations consisting of correlated Voff measurements:
RA=(VoffAbc+VoffAcb)−(VoffCab+VoffCba) (51)
RB=(VoffBca+VoffBac)−(VoffAbc+VoffAcb) (52)
RC=(VoffCab+VoffCba)−(VoffBca+VoffBac) (53)
(195) (ii) −30° Shifted Rotor Direction Function Calculations (2nd Embodiment (Δ))
(196) According to the second embodiment, the rotation direction functions (RA_Δ, RB_Δ, RC_Δ) shifted by −30° compared to the functions (RA, RB, RC) of the first embodiment are calculated with the following equations consisting of correlated Voff measurements:
RA_Δ=2.Math.(VoffAbc+VoffAcb)−(VoffCab+VoffCba)−(VoffBca+VoffBac) (54)
RB_Δ=2.Math.(VoffBca+VoffBac)−(VoffAbc+VoffAcb)−(VoffCab+VoffCba) (55)
RC_Δ=2.Math.(VoffCab+VoffCba)−(VoffBca+VoffBac)−(VoffAbc+VoffAcb) (56)
(197) (iii) Rotation Direction Detection Implementation
(198) At start-up, rotation direction of a BLDCM in motion is determined by calculating the disclosed (RA, RB, RC) RDFs, which evaluate to a positive or negative value depending on direction of rotation. This detection method is implemented during the initial rotor position operation as shown in the flowcharts in
(199) For the implementation of the 2nd embodiment, the function equations with the Δ suffix are used, for example RA_Δ instead of RA. The steps required to determine rotation direction are:
(200) 1. Calculate (RA, RB, RC) RDFs corresponding to commutation state determined during rotor position detection with RPFs and EPRFs. The sign of the RDF output determines forward or reverse rotation direction as shown in Table 7.
(201) TABLE-US-00009 TABLE 7 Rotation direction calculation with (RA, RB, RC) RDFs Rotor Position Commutation Forward Rotation Reverse Rotation Sector θ State RDF Sign Check RDF Sign Check 0-60° 0 RA > 0 RA < 0 60-120° 1 RC < 0 RC > 0 120-180° 2 RB > 0 RB < 0 180-240° 3 RA < 0 RA > 0 240-300° 4 RC > 0 RC < 0 300-360° 5 RB < 0 RB > 0
(202) (g) Initial Rotor Position Detection Extensions
(203) The disclosed initial rotor position detection method uses on-time and off-time open phase voltage measurements to detect inductance variations in BLDC motor phase coils. In practice, it is also possible to apply these disclosed methods to on-time and off-time inductance variation measurements obtained from phase current amplitude and rise time measurements in order to ascertain a BLDC motor's rotor position.
(204) Zero to Low Speed Sensorless BLDCM Operation
(205) The initial rotor position detection methods disclosed in previous sections provide accurate and precise starting rotor position information. To successfully drive a BLDC motor from standstill to high speed, a reliable sensorless operation in the zero to low speed range is required. This section discloses the RIMS inductance variation based commutation point functions (CPF) developed to provide accurate sensorless brushless commutation at zero and low motor speeds. They are combined with the previously disclosed rotor position functions (RPF, ERPF) and rotation direction functions (RDF) to deliver a highly robust sensorless operation at zero and low motor speeds in the presence of external load disturbances and with reversed rotor momentums.
(206) A PWM phase current control method consisting of three PWM time intervals is also disclosed. It allows effective injection of on-time and off-time open phase voltage measurement pulses required for the RIMS based commutation point detection. This PWM method is combined with a quasi Field Oriented Control (QFOC) algorithm developed to maximize output torque during 60° step sensorless operation.
(207) When motor speed is sufficiently high, sensorless operation switches over to the low to high speed sensorless commutation technique presented in the next section, as depicted in
(208) The key parts of the disclosed zero to low speed sensorless BLDCM operation are: PWM phase current control using three intervals (PWMT1-PWMT3) Commutation point detection functions (CPF) Rotor position and rotation direction functions (RPF, ERPF, RDF) 60° step sensorless commutation Quasi field oriented control (QFOC) CPFmax measurement used in rotor magnetic field strength and motor temperature measurement
(209) The operational flowchart of the disclosed zero to low speed sensorless BLDCM controller operation is illustrated in
(210) (a) PWM Phase Current Control
(211) The zero to low speed sensorless method of operation disclosed requires that the BLDCM phase driving currents are reduced to zero, as illustrated in
(212) 1. PWM T1: on-time phase current interval (T1), where the active phase windings (B+/C−) are connected to the supply (Vs) via the high-side power transistor (B_H) and to the ground (0V) rail via the low-side power transistor (C_L). A quasi FOC method, as described in later section of this disclosure, is also integrated to increase the BLDCM driving torque by maintaining a constant 90° torque angle between the BLDCM rotor and stator.
(213) 2. PWM T2: first off-time interval (T2) with slow phase current fall time, where both of the active phase coils (B and C) are connected to the ground (0V) rail via the low-side power transistors (B_L and C_L).
(214) 3. PWM T3: second off-time interval (T3) with fast phase current fall time down to zero, where only the B phase coil is actively connected to the ground (0V) rail via the low-side power transistor (B_L). The inactive high-side phase coil (C) is connected to the power supply rail (Vs) via the turn-off current conducted by the freewheeling diode of the high-side transistor (C_H), clipping the phase C voltage to Vs+Vd. The energy from this turn-off current is returned back to the power supply.
(215) Together these three PWM interval durations control the amplitude of the driving current which can be set depending on the BLDCM application. The control of PWM T1, PWM T2 and PWM T3 interval durations is not limited to any particular realization. For example, these PWM interval durations can be pre-calculated and stored in a look-up table in the controller memory, say to perform a simple open loop BLDCM motor speed control, or they can be calculated in real time to perform torque control or to achieve any other BLDCM controller driving optimization. An example of a simple method of PWM control, which has been tested to work well in practise, can be realized by setting the PWM T2 duration to zero and regulating the PWM T1 duration from 0% to 100% in order to control the level of the motor driving current. The PWM T1, PWM T2 and PWM T3 intervals of the disclosed zero to low speed sensorless method of operation are generated in the step shown in the operation flowchart in
(216) (b) On-Time and Off-Time Open Phase and Supply Voltage Measurements
(217) In this disclosure both the on-time and the off-time open phase voltage measurements are used to determine accurate sensorless operation commutation points using the developed commutation points functions. The same measurements are also used to determine the rotation direction of a BLDC motor. These on-time and off-time open phase voltage measurements are an important part of the zero to low speed sensorless BLDCM controller operation.
(218) As disclosed previously, the on-time and off-time phase supply voltages are also measured during this step to correlate the Von and Voff measurements and address the challenge of controller sensitivity to power supply voltage fluctuations. These steps in the zero to low speed controller operation are shown in the flowchart in
(219) (i) On-Time and Off-Time Voltage Measurement (1st Embodiment)
(220) The same method of on-time and off-time open phase voltage measurement, as illustrated in
(221) In this 1st controller embodiment, these on-time and off-time phase measurement pulses are inserted after the PWM (PWM T1, PWM T2, PWM T3) phase current control pulses, as shown in a practical BLDCM operation in
(222) TABLE-US-00010 TABLE 8 On-time and Off-time open phase and supply voltage measurements during zero to low speed sensorless BLDCM controller in the 1st embodiment On-time Off-time On-time Off-time Open Phase Open Phase Phase Supply Phase Supply Voltage Voltage Voltage Voltage Von0 Voff0 Vs0 Vsoff0 Von1 Voff1 Vs1 Vsoff1 Von2 Voff2 Vs2 Vsoff2 Von3 Voff3 Vs3 Vsoff3
(223) The Von measurements are next correlated to the first measurement (Von0) with following calculations:
Von0′=Von0 (57)
Von1′=Von1.Math.Vs0/Vs1 (58)
Von2′=Von2.Math.Vs0/Vs2 (59)
Von3′=Von3.Math.Vs0/Vs3 (60)
(224) Similarly, the Voff measurements are correlated to the first measurement (Voff0) with following calculations:
Voff0′=Voff0.Math.Vs0/Vsoff0 (61)
Voff1′=Von1.Math.Vs0/Vsoff1 (62)
Voff2′=Von2.Math.Vs0/Vsoff2 (63)
Voff3′=Von3.Math.Vs0/Vsoff3 (64)
(225) These correlated Von and Voff measurements are used in calculations of commutation point detection functions and rotation direction detection functions of the 1st embodiment of this invention, which are disclosed in the proceeding sections.
(226) (ii) On-Time and Off-Time Voltage Measurement (2nd Embodiment (Δ))
(227) In the 2nd controller embodiment up to six pairs of on-time and off-time phase measurement pulses are inserted after the PWM (PWM T1, PWM T2, PWM T3) phase current control pulses, as shown in a practical BLDCM operation in
(228) TABLE-US-00011 On-time Off-time On-time Off-time Open Phase Open Phase Phase Supply Phase Supply Voltage Voltage Voltage Voltage Von0 Voff0 Vs0 Vsoff0 Von1 Voff1 Vs1 Vsoff1 Von2 Voff2 Vs2 Vsoff2 Von3 Voff3 Vs3 Vsoff3 Von4 Voff4 Vs4 Vsoff4 Von5 Voff5 Vs5 Vsoff5
(229) Table 9: On-time and Off-time open phase and supply voltage measurements during zero to low speed sensorless BLDCM controller in the 2nd embodiment (Δ). The Von measurements are next correlated to the first measurement (Von0) with following calculations:
Von0′=Von0 (65)
Von1′=Von1.Math.Vs0/Vs1 (66)
Von2′=Von2.Math.Vs0/Vs2 (67)
Von3′=Von3.Math.Vs0/Vs3 (68)
Von4′=Von4.Math.Vs0/Vs4 (69)
Von5′=Von5.Math.Vs0/Vs5 (70)
(230) Similarly, the Voff measurements are correlated to the first measurement (Von0) with following calculations:
Voff0′=Voff0.Math.Vs0/Vsoff0 (71)
Voff1′=Von1.Math.Vs0/Vsoff1 (72)
Voff2′=Von2.Math.Vs0/Vsoff2 (73)
Voff3′=Von3.Math.Vs0/Vsoff3 (74)
Voff4′=Von4.Math.Vs0/Vsoff4 (75)
Voff5′=Von5.Math.Vs0/Vsoff5 (76)
(231) These correlated Von and Voff measurements are used in calculations of commutation point detection functions and rotation direction detection functions, in the 2nd embodiment of this invention, denoted with the A suffix, which are disclosed in the proceeding sections.
(232) (c) Commutation Point Detection
(233) This section discloses the developed commutation point functions (CPF), derived using the RIMS variation detection methods, from which six robust 60° step sensorless brushless commutation intervals are obtained. It is evident that BLDC motors are available in a variety of mechanical and winding configurations and as such they can have different RIMS variations properties. Consequently, two CPF detection embodiments are disclosed to cater for the differences in BLDC motors, with each having a further set of CPF detection method variations to address the challenge of sensorless controller operation with a broader range of BLDC motor applications. The required Von and Voff open phase winding measurements for CPF calculations are integrated into the PWM cycle of the zero to low speed sensorless BLDCM control method as shown in
(234) (i) Commutation Point Functions Using Von and Voff Measurements (1st Embodiment)
(235) In this first BLDCM controller embodiment, the correlated Von0 and Voff0 open phase measurements, obtained using the methods disclosed previously and as shown in
CPF0_Von_Voff=VoffCab−VonCab (0°≤θ<60°) (77)
CPF1_Von_Voff=VonBac−VoffBac (60°≤θ<120°) (78)
CPF2_Von_Voff=VoffAbc−VonAbc (120°≤θ<180°) (79)
CPF3_Von_Voff=VonCba−VoffCba (180°≤θ<240°) (80)
CPF4_Von_Voff=VoffBca−VonBca (240°≤θ<300°) (81)
CPF5_Von_Voff=VonAcb−VoffAcb (300°≤θ<360°) (82)
(236) A practical example of correlated Von and Voff open phase voltages measured in a BLDC motor is shown in
(237) This method of commutation point detection requires only the measurement of the first pulse shown in
(238) (ii) Commutation Point Functions Using Von and Von Measurements (1st Embodiment).
(239) In this first BLDCM controller embodiment, the correlated Von0 and Von1 open phase measurements, obtained using the methods disclosed previously and as shown in
CPF0_Von_Von=VonCba−VonCab (0°≤θ<60°) (83)
CPF1_Von_Von=VonBac−VonBca (60°≤θ<120°) (84)
CPF2_Von_Von=VonAcb−VonAbc (120°≤θ<180°) (85)
CPF3_Von_Von=VonCba−VonCab (180°≤θ<240°) (86)
CPF4_Von_Von=VonBac−VonBca (240°≤θ<300°) (87)
CPF5_Von_Von=VonAcb−VonAbc (300°≤θ<360°) (88)
(240) The commutation points are detected when the CPFs cross the zero voltage level (CPF≤0). This method requires two measurement pulses containing the Von0, Voff0 and Von1 and Voff1 measurement pairs, as shown in
(241) (iii) Commutation Point Functions Using Von and Voff Measurements (2nd Embodiment (Δ))
(242) During development it was found that with certain BLDCMs, such as delta winding configuration, it was not possible to obtain accurate commutation points using the CPFs (CPFX_Von_Voff, CPFX_Von_Von) disclosed in the first embodiment. In order to address the challenge of controller operation with the broadest possible range of BLDC motor applications, a further set of CPFs using combinations of Von and Voff open phase voltage measurements were developed. These work in conjunctions with the previously disclosed RPF and RDF equations of the 2nd embodiment, denoted with the Δ suffix.
(243) In this second BLDCM controller embodiment, the correlated Von0 and Voff0 open phase measurements, obtained using the methods disclosed previously and as shown in
CPF0_Von_Voff_Δ=VoffBac−VonBac (0°≤θ<60°) (89)
CPF1_Von_Voff_Δ=VonAbc−VoffAbc (60°≤θ<120°) (90)
CPF2_Von_Voff_Δ=VoffCba−VonCba (120°≤θ<180°) (91)
CPF3_Von_Voff_Δ=VonBca−VoffBca (180°≤θ<240°) (92)
CPF4_Von_Voff_Δ=VoffAcb−VonAcb (240°≤θ<300°) (93)
CPF5_Von_Voff_Δ=VonCab−VoffCab (300°≤θ<360°) (94)
(244) The commutation points are detected when the CPFs cross the zero voltage level (CPF≤0). This method of commutation point detection requires only the first measurement pulse containing the Von0 and Voff0 measurement, as shown in
(245) (iv) Commutation Point Functions Using Von and Von Measurements (2nd Embodiment (Δ))
(246) In this second BLDCM controller embodiment, the correlated Von0 and Von1 open phase measurements, obtained using the methods disclosed previously and as shown in
CPF0_Von_Von_Δ=VonBca−VonBac (0°≤θ<60°) (95)
CPF1_Von_Von_Δ=VonAbc−VonAcb (60°≤θ<120°) (96)
CPF2_Von_Von_Δ=VonCab−VonCba (120°≤θ<180°) (97)
CPF3_Von_Von_Δ=VonBca−VonBac (180°≤θ<240°) (98)
CPF4_Von_Von_Δ=VonAbc−VonAcb (240°≤θ<300°) (99)
CPF5_Von_Von_Δ=VonCab−VonCba (300°≤θ<360°) (100)
(247) The commutation points are detected when the CPFs cross the zero voltage level (CPF≤0). This method requires two measurement pulses containing the Von0, Voff0 and Von1 and Voff1 measurement pairs, as shown in
(248) (v) Commutation Point Function Detection Extensions
(249) 1. CPFs Obtained Using Other Combinations of Correlated Von and Voff Measurements
(250) The commutation point detection methods presented in the 1st and 2nd (Δ) controller embodiments are not only limited to the four disclosed CPFs (CPFX_Von_Voff, CPFX_Von_Von, CPFX_Von_Voff CPFX_Von_Von_Δ) which were found to work with a broad range of BLDC motors in practice. Any other combinations of correlated Von and Voff open phase winding voltage measurements can also be used with the disclosed method to obtain accurate commutation points to cater for special BLDC motor applications, for example using Von and Voff measurement combinations of the form:
VonXxx−VoffXxx
VoffXxx−VoffXxx
VonXxx−VonXxx
(251) In practice, it was found that the selection of the most suitable CPFs is dependent on the characteristics of the Von and Voff curves measured in a BLDC motor, for example as shown in
(252) 2. CPFs Obtained from Von or Voff Measurements in Consecutive PWM T1-T3 Cycles
(253) Furthermore, as an extension to the disclosed commutation point detection methods of the 1.sup.st and 2.sup.nd embodiments, which use Von and Von measurements (CPFX_Von_Von) in a single PWM cycle, it is in practice also possible to combine Von and Von measurements from two consecutive PWM cycles for use in CPF calculations, as shown in
CPFX_Von_Von[1]=Von[1]−Von[0]
CPFX_Von_Von[2]=Von[2]−Von[1]
CPFX_Von_Von[3]=Von[3]−Von[2]
(254) During commutation point detection with Von and Von measurements this has the advantage of reducing the number of required measurement pulses from two to one in each PWM T1-T3 cycle, which is then equivalent to the single pulse measurement method in the disclosed CPF calculation employing Von and Voff measurements (CPFX_Von_Voff). The same method can also be applied to CPFs obtained using only Voff measurements.
(255) 3. Commutation Points Obtained from Absolute Rotor Position (θ) Calculations
(256) As a further variation to the disclosed commutation point detection methods for both the 1.sup.st and 2.sup.nd (Δ) embodiments, it is in practise also possible to calculate accurate commutation points for reliable sensorless brushless motor operation by using rotor position information obtained from the previously disclosed method involving the calculation of absolute rotor position (θ) in equation (50). As illustrated in the operation flowchart shown in
(257) The necessary commutation points (CPF0-5) are then obtained directly from the calculated absolute rotor position, by checking the value of 0:
CPF0: θ>=60°
CPF1: θ>=120°
CPF2: θ>=180°
CPF3: θ>=240°
CPF4: θ>=300°
CPF5: θ>=360°
(258) (d) Rotation Direction Detection
(259) To address the challenge of robust sensorless BLDCM controller operation at and near zero speeds and in the presence of external load disturbances and with reversed rotor momentum, a method for determining the rotation direction during the zero to low speed sensorless BLDCM operation is disclosed in this section. The rotation detection employs the calculation of the RPFs, ERPFs and RDFs, presented previously in the initial rotor position detection method disclosure of this controller invention.
(260) Two rotation direction calculation embodiments are also disclosed, with the 2.sup.nd embodiment denoted with the Δ suffix. For the implementation of the 2.sup.nd embodiment, the function equations with the Δ suffix are used, for example RA_Δ instead of RA.
(261) (i) Von and Voff Measurement
(262) The required Von and Voff open phase winding measurements for rotation direction calculation are inserted after the completion of the PWM T1-PWM T3 stream, as shown in
(263) (ii) Rotation Direction Detection with RDFs and RPFs
(264) The rotation direction detection truth table utilizing the RPF and RDF calculations for all six commutation intervals (0-360°) during sensorless operation is shown in Table 10. For example, if during forward operation forward rotation is detected, then the commutation state is incremented. Conversely, if reverse rotation is detected then the commutation state is decreased. This procedure is summarized in the operation flowchart in
(265) TABLE-US-00012 TABLE 10 Rotation direction detection during zero to low speed sensorless BLDCM operation with RDFs (RA, RB, RC) and RPFs (PA_30, PB_30, PC_30) Rotor Position Commutation Forward Rotation Reverse Rotation Sector θ State Detection Condition Detection Condition 0-60° 0 RA ≥ 0 || PA_30 ≥ 0 !(RA ≥ 0 || PA_30 ≥ 0) 60-120° 1 RC ≤ 0 || PC_30 ≤ 0 !(RC ≤ 0 || PC_30 ≤ 0) 120-180° 2 RB ≥ 0 || PB_30 ≥ 0 !(RB ≥ 0 || PB_30 ≥ 0) 180-240° 3 RA ≤ 0 || PA_30 ≤ 0 !(RA ≤ 0 || PA_30 ≤ 0) 240-300° 4 RC ≥ 0 || PC_30 ≥ 0 !(RC ≥ 0 || PC_30 ≥ 0) 300-360° 5 RB ≤ 0 || PB_30 ≤ 0 !(RB ≤ 0 || PB_30 ≤ 0)
(266) (iii) Rotation Direction Detection with RDFs and ERPFs
(267) The rotation direction detection truth table utilizing the ERPF and RDF calculations for all six commutation intervals (0-360°) during sensorless operation is shown in Table 11. It is valid for any of the CPF detection methods. The operation procedure is the same as with the RPF calculations, as shown in the flowchart in
(268) TABLE-US-00013 TABLE 11 Rotation direction detection during zero to low speed sensorless BLDCM operation with RDFs (RA, RB, RC) and ERPFs (EPA0_30, EPC1_30, EPB2_30, EPA3_30, EPC4_30, EPB5_30) Rotor Com- Position mutation Forward Rotation Reverse Rotation Sector θ State Detection Condition Detection Condition 0-60° 0 RA ≥ 0 || EPA0_30 ≥ 0 !(RA ≥ 0 || EPA0_30 ≥ 0) 60-120° 1 RC ≤ 0 || EPC1_30 ≤ 0 !(RC ≤ 0 || EPC1_30 ≤ 0) 120-180° 2 RB ≥ 0 || EPB2_30 ≥ 0 !(RB ≥ 0 || EPB2_30 ≥ 0) 180-240° 3 RA ≤ 0 || EPA3_30 ≤ 0 !(RA ≤ 0 || EPA3_30 ≤ 0) 240-300° 4 RC ≥ 0 || EPC4_30 ≥ 0 !(RC ≥ 0 || EPC4_30 ≥ 0) 300-360° 5 RB ≤ 0 || EPB5_30 ≤ 0 !(RB ≤ 0 || EPB5_30 ≤ 0)
(269) (iv) Rotation Direction Detection Extensions
(270) Depending on the BLDCM application, the disclosed forward/reverse rotation direction detection check can also be implemented using only RDFs (RA, RB, RC) calculations or only RPFs (PA_30, PB_30, PC_30)/ERPFs (EPA0_30, EPC1_30, EPB2_30, EPA3_30, EPC4_30, EPB5_30) calculations in order to reduce controller computation complexity.
(271) Furthermore, it is also possible to obtain accurate rotation direction information by utilizing the absolute rotor position (θ) calculation disclosed previously in equation (50). This requires the six pairs of correlated on-time (VonXxx) and off-time (VoffXxx) open phase measurements, as shown in the practical example in
Forward Rotation: θ[n+1]>θ[n]
Reverse Rotation: θ[n+1]<θ[n]
(272) (v) Rotation Direction Detection Extensions Using Enhanced Rotor Position Functions
(273) In application with certain types of BLDC motors, such internal permanent magnet (IPM) motors, it was found that the calculation order of the +30° shifted enhanced rotor position functions (EPA0_30, EPC1_30, EPB2_30, EPA3_30, EPC4_30, EPB5_30) for both the 1st and 2nd (Δ) embodiments had to be modified, in order to obtain the same effect. This is achieved by swapping the enhanced rotor position function calculations with the same phase coil pairs (eg. EPAx_30, EPBx_30, EPCx_30) as outlined below:
EPA0x_30=EPA3x_30 (0°≤θ<60°)
EPC1x_30=EPC4x_30 (60°≤θ<120°)
EPB2x_30=EPB5x_30 (120°≤θ<180°)
EPA3x_30=EPA0x_30 (180°≤θ<240°)
EPC4x_30=EPC1x_30 (240°≤θ<300°)
EPB5x_30=EPB2x_30 (300°≤θ<360°)
(274) Where: x={‘ ’, _Δ}
(275) Thus, this calculation swap applies to all of the following disclosed enhanced rotor position function calculations shifted by +30°, of the 1st and 2nd (Δ) embodiments: (EPA0_30, EPC1_30, EPB2_30, EPA3_30, EPC4_30, EPB5_30) (EPA0_Δ_30, EPC1_Δ_30, EPB2_Δ_30, EPA3_Δ_30, EPC4_Δ_30, EPB5_Δ_30)
(276) (e) Sensorless BLDCM Commutation
(277) The disclosed sensorless commutation point detection methods deliver decisive and robust commutation points to control the BLDCM stator magnetic field in 60° steps. The strength of the stator magnetic field is determined by the magnitude of the BLDCM phase currents, which are regulated with the disclosed PWM method consisting of three distinct intervals (PWM T1, PWM T2, PWM T3). This step is illustrated in operation flowchart in
(278) (i) Sensorless Commutation (1st Embodiment)
(279) The active phase coils in each commutation state during PWM T1, PWM T2 and PWM T3 intervals in the 1st embodiment of this invention are summarized in Table 12, Table 13, Table 14 respectively. The corresponding commutation point function detection and the phase winding power transistor switching sequence for each of the six commutation steps is shown in
(280) TABLE-US-00014 TABLE 12 PWM T1 interval phase coil switching states in the 1st embodiment Rotor Position Commutation Top Active Bottom Active Sector θ State Phase Coil Phase Coil 0-60° 0 A+ C− 60-120° 1 B+ C− 120-180° 2 B+ A− 180-240° 3 C+ A− 240-300° 4 C+ B− 300-360° 5 A+ B−
(281) TABLE-US-00015 TABLE 13 PWM T2 interval phase coil switching states in the .sup.st embodiment Rotor Position Commutation Top Active Bottom Active Sector θ State Phase Coil Phase Coil 0-60° 0 — C−, A− 60-120° 1 — C−, B− 120-180° 2 — A−, B− 180-240° 3 — A−, C− 240-300° 4 — B−, C− 300-360° 5 — B−, A−
(282) TABLE-US-00016 TABLE 14 PWM T3 interval phase coil switching states in the 1st embodiment Rotor Position Commutation Top Active Bottom Active Sector θ State Phase Coil Phase Coil 0-60° 0 — A− 60-120° 1 — B− 120-180° 2 — B− 180-240° 3 — C− 240-300° 4 — C− 300-360° 5 — A−
(283) (ii) Sensorless Commutation (2nd Embodiment (Δ))
(284) The active phase coils in each commutation state during PWM T1, PWM T2 and PWM T3 intervals in the 2nd embodiment (Δ) are summarized in Table 15, Table 16, Table 17 respectively. The corresponding commutation point function detection and the phase winding power transistor switching sequence for each of the six commutation steps is shown in
(285) TABLE-US-00017 TABLE 15 PWM T1 interval phase coil switching states in the 2nd embodiment (Δ) Rotor Position Commutation Top Active Bottom Active Sector θ State Phase Coil Phase Coil 0-60° 0 A+ B−, C− 60-120° 1 A+, B+ C− 120-180° 2 B+ A−, C− 180-240° 3 B+, C+ A− 240-300° 4 C+ A−, B− 300-360° 5 A+, C+ B−
(286) TABLE-US-00018 TABLE 16 PWM T2 interval phase coil switching states in the 2nd embodiment (Δ) Rotor Position Commutation Top Active Bottom Active Sector θ State Phase Coil Phase Coil 0-60° 0 — C−, A−, B− 60-120° 1 — C−, B−, A− 120-180° 2 — A−, B−, C− 180-240° 3 — A−, C−, B− 240-300° 4 — B−, C−, A− 300-360° 5 — B−, A−, C−
(287) TABLE-US-00019 TABLE 17 PWM T3 interval phase coil switching states in the 2nd embodiment (Δ) Rotor Position Commutation Top Active Bottom Active Sector θ State Phase Coil Phase Coil 0-60° 0 — A− 60-120° 1 — B−, A− 120-180° 2 — B− 180-240° 3 — C−, B− 240-300° 4 — C− 300-360° 5 — A−, C−
(288) (f) Quasi Field Oriented Control
(289) In the previously disclosed method of sensorless commutation, the stator magnetic field remains fixed while the rotor turns through an angle of 60°. The relative angle between stator and rotor magnetic fields thus changes from 120° to 60°. The maximum BLDC motor torque is produced when the angle between the stator and rotor magnetic fields is 90°.
(290) This disclosure describes a quasi FOC method used to estimate continuous rotor position with which a constant 90° torque angle can be produced in a BLDCM to address this challenge and improve performance. It uses the information contained in the previously disclosed CPFs which in general have been found to exhibit sinusoidal waveform behaviour, as shown in the BLDCM example in
(291) An inverse sine operation can be used to calculate the continuous rotor angle (φ) from the CPFs with the calculations given below, where CPFmax is the peak amplitude of the CPF in each 60° rotor position sector:
φ=sin.sup.−1(CPF/CPFmax)/3 (0°≤θ<30°) (101)
φ=60−sin.sup.−1(CPF/CPFmax)/3 (30°≤θ≤60°) (102)
(292) The estimated continuous rotor angle (φ) is then used to modulate the third phase coil in each 60° commutation state. This disclosed QFOC calculation step is shown in the operation flowchart in
(293) It should be noted that in practical applications, CPFs which do not exhibit sinusoidal waveform behaviour can also utilise other possible calculation methods or functions, for example, such as higher-order polynomial functions or piecewise linear functions, in order to deduce continuous rotor angle (φ) from the measured CPF values, as required by the disclosed QFOC method of operation.
(294) (i) QFOC (1st Embodiment)
(295) To implement the QFOC operation in the 1st embodiment of this invention, the active phase coil switching states during PWM T1 interval operation are modified as shown in Table 18. The QFOC modulated phase coil is switched in a complimentary manner between the supply voltage (Vs) and ground rail (0V), with the duty cycle proportional to the estimated rotor position angle (φ). The corresponding commutation point function detection and the phase winding power transistor switching sequence employing QFOC for each of the six commutation steps is shown in
(296) TABLE-US-00020 TABLE 18 PWM T1 interval quasi FOC operation phase coil switching states in the 1st embodiment Rotor Top Bottom QFOC Position Commutation Active Active φ Modulated Sector θ State Phase Coil Phase Coil Phase Coil 0 .fwdarw. 60° 0 A+ C− B+− (0 .fwdarw. 100%) 60 .fwdarw. 120° 1 B+ C− A+− (100 .fwdarw. 0%) 120 .fwdarw. 180° 2 B+ A− C+− (0 .fwdarw. 100%) 180 .fwdarw. 240° 3 C+ A− B+− (100 .fwdarw. 0%) 240 .fwdarw. 300° 4 C+ B− A+− (0 .fwdarw. 100%) 300 .fwdarw. 360° 5 A+ B− C+− (100 .fwdarw. 0%)
(297) (ii) QFOC (2nd Embodiment (Δ))
(298) To implement the QFOC operation in the 2nd embodiment, the active phase coil switching states during PWM T1 interval operation are modified as shown in Table 19. In this embodiment, the QFOC modulation uses either the bottom or top side power transistor to switch the phase coil, with a duty cycle proportional to the estimated rotor position angle (φ). The corresponding commutation point function detection and the phase winding power transistor switching sequence employing QFOC for each of the six commutation steps is shown in
(299) TABLE-US-00021 TABLE 19 PWM T1 interval quasi FOC operation phase coil switching states in the 2nd embodiment (Δ) Rotor Top Bottom QFOC Position Commutation Active Active φ Modulated Sector θ State Phase Coil Phase Coil Phase Coil 0 .fwdarw. 30° 0 A+ B− C− (0 .fwdarw. 100%) 30 .fwdarw. 60° 0 A+ C− B− (100 .fwdarw. 0%) 60 .fwdarw. 90° 1 A+ C− B+ (0 .fwdarw. 100%) 90 .fwdarw. 120° 1 B+ C− A+ (100 .fwdarw. 0%) 120 .fwdarw. 150° 2 B+ C− A− (0 .fwdarw. 100%) 150 .fwdarw. 180° 2 B+ A− C− (100 .fwdarw. 0%) 180 .fwdarw. 210° 3 B+ A− C+ (0 .fwdarw. 100%) 210 .fwdarw. 240° 3 C+ A− B+ (100 .fwdarw. 0%) 240 .fwdarw. 270° 4 C+ A− B− (0 .fwdarw. 100%) 270 .fwdarw. 300° 4 C+ B− A− (100 .fwdarw. 0%) 300 .fwdarw. 330° 5 C+ B− A+ (0 .fwdarw. 100%) 330 .fwdarw. 360° 5 A+ B− C+ (100 .fwdarw. 0%)
(300) Rotor Magnetic Field Strength Measurement
(301) The developed commutation point functions (CPF) disclosed in the zero to low speed sensorless BLDCM operation present an opportunity to ascertain the relative rotor magnetic field strength measurement and addresses the challenge of BLDCM rotor performance monitoring in real time and over its entire life time of operation. During the development, it was found that the peaks (CPFmax) of the calculated CPFs are directly proportional to the rotor magnetic field strength in a BLDCM. Thus the greater the maximum CPF value, the greater is the resultant rotor magnetic field strength, as shown in
(302) To obtain a useful indicator of the relative rotor magnetic field strength for use in BLDCM applications, the disclosed method involves the calculation of maximum CPF value (CPFmax) which is independent of the power supply voltage (Vs), as shown below:
CPFmaxR=CPFmax/Vs (103)
(303) The resultant magnetic field strength saliency ratio (CPFmaxR) curves are illustrated in
(304) Sensorless Motor Temperature Measurement
(305) A method of motor winding temperature measurement using phase current and voltage measurements to calculate copper phase winding resistance, which is proportional to its temperature, is disclosed in U.S. Pat. No. 4,083,001. High accuracy phase current measurement requirement however, increases controller's electronic circuit complexity, size and cost.
(306) As an extension to the previously disclosed rotor magnetic field strength measurement, it was found that the same CPFmaxR measurement can also be used to indicate the relative motor temperature during real time operation. Those skilled in this art can appreciate that this phenomenon occurs because the magnetic field strength of a rotor is directly proportional its temperature.
(307) This relationship between the motor temperature and the rotor magnetic field strength saliency ratio (CPFmaxR) has been found to be linear for common BLDC motors tested. In practice, BLDC motors consisting of different electrical and mechanical construction and rotor magnetic field strength, also exhibit different CPFmaxR profiles. In the disclosed sensorless motor temperature measurement method, precise motor temperature measurements are obtained by tuning the individual BLDC motor temperature and CPFmaxR profiles. This is achieved by first measuring two pairs of motor temperature and magnetic field strength saliency ratio (CPFmaxR) measurements at cold (CPFmaxR_T1, T1) and hot (CPFmaxR_T2, T2) motor temperatures, as illustrated in
T=T2+(T1−T2).Math.(CPFmaxR−CPFmaxR_T2)/(CPFmaxR_T1−CPFmaxR_T2) (104)
(308) The disclosed embodiment of sensorless motor temperature measurement method, using a straight line approximation, is illustrated in
(309) The disclosed method can thus be used to monitor a BLDC motor temperature during operation without any additional temperature sensing hardware and electronic circuitry, saving system cost and increasing operational reliability. This calculation step during the zero to low speed sensorless BLDCM operation is shown in the flowchart in
(310) Low to High Speed Sensorless BLDCM Operation
(311) The RIMS saliency based initial rotor position detection and the zero to low speed sensorless BLDCM commutation technique disclosed in previous sections provide faultless motor starts and operation at and near zero speeds. This section discloses a second sensorless BLDCM controller method developed to operate BLDCMs at low and high motor speeds. This part of the hybrid controller approach is illustrated in
(312) The developed sensorless brushless commutation point detection (CPD) method employs conventional BEMF open phase voltage measurements to operate BLDCMs at very high speeds. To extend sensorless BLDCM operation into the low speed region and close to zero speeds, the disclosed method also utilizes inductance variations voltage measurements due to RIMS saliency to detect accurate rotor position and commutation points.
(313) Several prior art solutions which utilize inductance variations to detect sensorless commutation points in BLDCM applications are known. CH698071 presents a technique of measuring the inductance variation voltage at the open-phase terminals during PWM on-time and off-time motor driving switching intervals. From these +30° advanced timing motor commutation points are obtained. However, these have the significant disadvantage of reduced output motor torque and efficiency. U.S. Pat. Nos. 7,768,226 and 9,391,553 use inductance variation voltage measurements to determine the commutation points, however these methods require additional hardware circuitry as well as a fourth motor neutral point connection which limits the range of possible applications only to wye configuration BLDC motors. US20140062364 describes another similar method requiring motor's neutral point information to obtain sensorless commutation points, however in this method an external circuit is needed to simulate a virtual motor neutral point. U.S. Pat. Nos. 8,552,671 and 8,593,098 rely on changes in phase current measurements due to inductance variations to determine commutation points, however these methods require expensive and accurate current sensing measurement circuitry. U.S. Pat. No. 8,773,060 describes a technique of commutation point detection utilizing voltage measurements due to inductance variations and addressing the challenge of operation at higher BLDC motor temperatures. However, this method requires motor temperature measurement using external temperature sensors built into the motor and associated electronic circuitry, increasing the system costs. It also requires more complicated learning algorithms to adjust the commutation point detection voltage thresholds for BLDC motors operating at different temperatures.
(314) In contrast to prior art methods, the disclosed low to high speed sensorless BLDCM method uses the peaks of commutation point functions (CPFmax), measured during the previously disclosed zero to low speed sensorless BLDCM operation, to set the optimum commutation point detection voltage threshold levels. The resultant commutation points are optimum at all motor temperatures and exhibit quasi 0° commutation timing for maximum motor driving torque. To address the challenge of increased controller efficiency during PWM off-time switching, a software controlled PWM synchronous rectifier is also integrated into the PWM switching sequence to control optimum synchronous rectifier on-time duration. Standard PWM switching technique is used to control motor driving phase currents. The developed PWM switching sequence when B+/C− coil pair is energized is illustrated in
(315) The key innovations of the disclosed sensorless BLDC motor operation include: 1) A Robust and simple commutation point detection (CPD) method implemented during PWM on-time (PWM TON) switching interval, employing inductance voltage variations due to RIMS saliency and maximum values of the commutation point functions (CPFmax) measured during zero to low speed sensorless BLDCM operation; 2) Optimum commutation point detection voltage threshold setting for each BLDCM application, with quasi 0° commutation timing and automatic CPD voltage threshold compensation for motor temperature variations; 3) Sensorless BLDCM operation near zero speeds, low and high motor speeds; 4) Software controlled PWM synchronous rectification (PWM TSR), optimally controlled during PWM off-time (PWM TOFF) switching interval, to increase controller efficiency; 5) Reduced controller hardware complexity and cost.
(316) The operation flowchart of the disclosed low to high speed sensorless BLDCM controller operation is illustrated in
(317) (a) Quasi 0° Timing Commutation Point Detection at Low Speeds
(318) Conventional sensorless brushless operation utilizes BEMF voltages during PWM on-time to detect commutation points when the measured open phase voltage crosses the half way point of the supply voltage rail (Vs/2). Compared to the ideal 0° timing commutation intervals, this point occurs at +30° advanced timing, resulting in reduced motor torque and efficiency. To overcome this challenge, the sensorless BLDCM commutation point detection method disclosed employs offset voltage thresholds with which quasi 0° commutation timing points are attained.
(319) (i) PWM On-Time Open Phase and Supply Voltage Measurement
(320) The first two steps of the disclosed method involve energizing a pair of phase coils and measuring the open phase and supply phase voltage (Vs), as shown in the PWM switching example in
(321) (ii) PWM On-Time Commutation Point Detection
(322) RIMS saliency properties are utilized next to detect accurate sensorless commutation points during the PWM on-time switching interval at low motor speeds. An example of this is shown in
(323) 1. Minimum PWM on-time duration is limited to ˜15 us. This ensures that a phase current pulse with sufficient amplitude is injected into the energized coils to evoke RIMS saliency effects in the open phase voltage measurements required for commutation point detection
(324) 2. At 100% PWM duty cycle of operation, RIMS saliency effects are not measurable. Thus in practice, during operation at near zero and low speed the maximum PWM on-time duty cycle is restricted to ˜95%
(325) To obtain reliable sensorless BLDCM operation commutation points, a voltage threshold offset corresponding to a factor of the maximum magnitude of the trough or peak voltage (CPFmax/2) is added to Vs/2, as illustrated in
(326) (iii) PWM On-Time H.fwdarw.L and L.fwdarw.H CPD
(327) The following two commutation point detection thresholds are calculated during PWM on-time interval for each open phase voltage measurement in the commutation sequence.
(328) H.fwdarw.L Open Phase Voltage CPD Threshold: Vs/2−N.Math.CPFmax
(329) L.fwdarw.H Open Phase Voltage CPD Threshold: Vs/2+N.Math.CPFmax
(330) Where: CPFmax is the maximum value of the commutation point function measured during zero to low speed sensorless BLDCM operation. It is dependent on BLDC motor temperature, as disclosed previously above. The scaling factor, N is used to adjust the voltage threshold offset level and the resultant quasi 0° commutation timing point. Values in the range of 0.25−0.5 were tested to work well in practice. Vs is the supply voltage measured during PWM on-time
(331) (iv) PWM On-Time Commutation Point Detection Calculations
(332) In one complete BLDCM electrical rotor cycle, the disclosed commutation point detection calculations during PWM on-time interval for the six commutation states corresponding are given by:
CP0=Vs.Math.(½+N.Math.CPFmaxR)−Vb (0°≤θ<60°) (105)
CP1=Va−Vs.Math.(½−N.Math.CPFmaxR) (60°≤θ<120°) (106)
CP2=Vs.Math.(½+N.Math.CPFmaxR)−Vc (120°≤θ<180°) (107)
CP3=Vb−Vs.Math.(½.Math.N.Math.CPFmaxR) (180°≤θ<240°) (108)
CP4=Vs.Math.(½+N.Math.CPFmaxR)−Va (240°≤θ<300°) (109)
CP5=Vc−Vs.Math.(½−N.Math.CPFmaxR) (300°≤θ<360°) (110)
(333) Where: CPFmaxR is the supply voltage independent ratio value (CPFmax/Vs), disclosed previously in equation (103); Va, Vb, Vc are the open phase voltages measured during PWM on-time switching intervals (PWM TON) in the corresponding commutation state interval. Vs is the supply voltage measured during PWM on-time. The commutation point is detected when the calculated CP≤0. This step is illustrated in the operational flowchart in
(334) (b) Synchronous Rectification During PWM Off-Time
(335) Some prior art solutions employing synchronous rectification techniques in brushed motor control applications are disclosed in U.S. Pat. Nos. 6,384,555 and 6,956,359. However, these solutions require additional hardware circuitry to perform this function and thus have higher controller cost. Furthermore, it is likely that no prior art solution has specifically addressed the challenge of synchronous rectification in sensorless brushless BLDC motor applications.
(336) The disclosed low to high speed sensorless operation integrates a software controlled synchronous rectification technique during PWM off-time to reduce controller power loses. A practical example of the realized synchronous rectification operation is shown in
(337) (c) Commutation Point Detection at High Speeds
(338) At higher speeds of operation, the disclosed commutation point detection during PWM on-time interval automatically reverts to operation equivalent to the conventional sensorless brushless commutation point detection utilizing BEMF voltage measurements. The commutation timing increases from the quasi 0° to 30°, which can be retarded back to 0° using time delay techniques that are well know to those skilled in this art.
(339) (d) Sensorless BLDCM Commutation and Synchronous Rectifier Phase Coils Switching States
(340) The active phase coils in each commutation state during PWM T.sub.ON and PWM T.sub.SR switching intervals are equivalent to the previously disclosed states in Table 12 and Table 13 respectively. The CPs and corresponding PWM on-time and synchronous rectifier phase coil switching states for one complete electrical revolution in a BLDC motor are illustrated in
(341) (e) Low to High Speed Sensorless BLDCM Operation Extensions
(342) The disclosed low to high speed sensorless BLDCM operation has been presented for the case of high-side PWM switching topology, commonly used with bootstrapped transistor gate drivers. However, the disclosed technique of commutation point detection and software controlled synchronous rectification can be applied to any other PWM switching topology, such as for example, low-side PWM switching and alternate low-high side PWM switching, in order to make optimum use of the implemented controller hardware circuitry.
(343) Software Controlled PWM Synchronous Rectification
(344) Synchronous rectification (SR) is a technique used to improve controller output efficiency during PWM switching operation with inductive loads. Generally, this operation is implemented with external hardware circuitry which increases overall system cost, as for example disclosed in U.S. Pat. Nos. 6,384,555, 6,396,250, 6,861,826 and 6,956,359.
(345) This section discloses a synchronous rectification method implemented entirely with a software feedback control algorithm, which does not require any additional hardware. It is integrated into the previously disclosed low to high speed sensorless BDLCM operation and the regenerative BLDC motor braking disclosed in proceeding section.
(346) (a) Synchronous Rectification Background
(347) In inductive load PWM switching applications, such as BLDC motors, the phase current which rises exponentially during PWM on-time interval continues to flow in the PWM off-time interval. Normally this current flow is maintained via a freewheeling diode, either integrated into the semiconductor switch, such as a MOSFET or IGBT, or by a discrete diode device. Improvement in efficiency can be made when the freewheeling diode is replaced with a lower resistance semiconductor device to carry the phase current during PWM off-time. The timing of this operation is important and must be synchronized with the PWM off-time, hence this process is called “synchronous rectification”. The following section outlines the difference between these two methods commonly applied in the prior art.
(348) (i) Freewheeling Diode Rectifier
(349)
P.sub.d_FW=V.sub.d.Math.I.sub.OFF[W] (111)
(350) This power loss, illustrated in
(351) (ii) Synchronous Rectifier
(352) The phase voltage and current waveforms during PWM off-time synchronous rectifier operation are illustrated in
P.sub.d_SR=R.sub.ds_on.Math.I.sub.OFF.sup.2[W] (112)
(353) With modern semiconductor switches, such as MOSFETs, exhibiting very low solid state on-resistance values (<1 mΩ), the power loss during the synchronous rectifier operation, illustrated in
(354) (b) Software Synchronous Rectifier Feedback Controller
(355) This section discloses the synchronous rectifier feedback control algorithm implemented in software, which was developed to eliminate additional hardware circuitry found in the prior art. The objective of the feedback controller is to maintain a zero phase current at the end of the SR period during the PWM off-time interval. This produces the optimal synchronous rectifier duration synonymous with the prior art external hardware solutions. Depending on the length of SR duration (T.sub.SR) with respect to the PWM off-time interval duration (T.sub.OFF) two distinct SR feedback control measurement techniques are employed. The disclosed synchronous operation flowchart is presented in
(356) (c) SR Feedback Voltage (VFB) Measurement
(357) The disclosed software controlled synchronous rectifier requires the measurement of the feedback voltage (V.sub.FB) at the end of the synchronous rectification duration (T.sub.SR). This task is shown in operation flowchart in
(358) (d) SR Feedback Controller with T.sub.SR<T.sub.OFF
(359) In operation at low inductive loads, for example when a BLDCM has reached higher speeds, the phase currents may have sufficient time to decay to zero before reaching the end of PWM off-time interval (T.sub.OFF). During this discontinuous phase current mode of PWM operation the SR duration (T.sub.SR) is less than the entire PWM off-time duration (T.sub.OFF). Illustrated in
(360) (i) SR I.sub.Fb>0 (V.sub.Fb>0)
(361) The phase voltage and current waveforms and power transistor switching states during SR feedback control operation when I.sub.FB>0 are shown in
(362) (ii) SR I.sub.FB<0 (V.sub.FB<0)
(363) The case when the SR duration (T.sub.SR) is too long compared to the optimal duration, resulting in “over” rectification, is illustrated in
(364) (iii) SR Feedback Control Law with T.sub.SR<T.sub.OFF
(365) The SR feedback control law during SR operation with T.sub.SR<T.sub.OFF is given by: If V.sub.FB>0 then increase T.sub.SR duration If V.sub.FB≤0 then decrease T.sub.SR duration
(366) In practice the T.sub.SR duration can be regulated using simple increment and decrement algorithms, with the speed of response weighed according to application requirements. One example implemented is given by:
T.sub.SR[n+1]=T.sub.SR[n]±ΔT.sub.SR.Math.W (113)
(367) Where: ΔT.sub.SR is the increment/decrement step; W is the increment/decrement weight used to adjust the feedback controller response speed. More advanced control techniques, such as PID can also be implemented to regulate the SR feedback control loop.
(368) (e) SR Feedback Controller with T.sub.SR=T.sub.OFF
(369) During operation at high inductive loads, typically when a BLDCM is starting and the speed is low, the phase currents do not have sufficient time to decay to zero before reaching the end of the PWM off-time interval (T.sub.OFF). This results in a continuous phase current mode of PWM operation. In this case, the SR duration (T.sub.SR) reaches its maximum possible value, equal to the entire PWM off-time duration (T.sub.OFF). Similar to operation when T.sub.SR<T.sub.OFF disclosed in previous section, two distinct SR feedback measurements illustrated in
(370) (i) SR I.sub.Fb>0 (V.sub.Fb<0)
(371) The phase voltage and current waveforms and power transistor switching states during SR feedback control operation when I.sub.FB>0, are shown in
(372) (ii) SR I.sub.Fb<0 (V.sub.Fb>0)
(373) The case when the SR period is too long, resulting in over-rectification, is illustrated in
(374) (iii) SR Feedback Control Law with T.sub.SR=T.sub.OFF
(375) The SR feedback control law during SR operation with T.sub.SR=T.sub.OFF is given by: If V.sub.FB≤0 then increase T.sub.SR duration If V.sub.FB>0 then decrease T.sub.SR duration
(376) T.sub.SR duration is regulated using the same methods previously outlined for the SR feedback control law with T.sub.SR<T.sub.OFF.
(377) (f) Software Synchronous Rectifier Extensions
(378) The disclosed software controlled synchronous rectification has been presented for the case of high-side PWM switching topology, commonly used with bootstrapped transistor gate drivers. However, the disclosed technique can be applied to any other PWM switching topology, such as for example, low-side PWM switching and alternate low-high side PWM switching.
(379) It is also evident that the disclosed software controlled synchronous rectification method can be applied to any inductive PWM switching application, such as for example, DC/DC power supplies, during normal motor driving and motor braking operation of brushed motors, sensorless brushless motors and sensored brushless motors.
(380) Regenerative Motor Braking with Synchronous Rectification
(381) This section discloses the regenerative motor braking method for sensorless BLDC motor applications, which is integrated into this controller invention as shown in
(382) The disclosed method address the following key challenge areas of regenerative motor braking applications compared to the prior art: 1) Increased controller efficiency, resulting in reduced power and heating losses during regenerative motor braking; 2) Increased level of energy returned back to the power source, resulting in increased time of operation in applications such as battery powered applications; 3) Reduced controller hardware complexity and cost, requiring no additional circuitry to increase regenerative motor braking performance.
(383) (a) Commutation Point Detection
(384) Conventional sensorless brushless commutation point detection methods, known to those skilled in this art, utilizing BEMF voltage measurements are employed during the regenerative BLDC motor braking. These can be applied during PWM on-time and off-time intervals shown in
(385) (i) PWM On-Time CPD
(386) As illustrated in the example of
CP0_Ton=−Vb (330°≤θ<30°) (114)
CP1_Ton=Va (30°≤θ<90°) (115)
CP2_Ton=−Vc (90°≤θ<150°) (116)
CP3_Ton=Vb (150°≤θ<210°) (117)
CP4_Ton=−Va (210°≤θ<270°) (118)
CP5_Ton=Vc (270°≤θ<330°) (119)
(387) The commutation point is detected when the calculated CP≤0. This step is illustrated in the regenerative motor braking operation flowchart in
(388) (ii) PWM Off-Time CPD with Synchronous Rectifier
(389) During this PWM off-time interval, the active phase coils (B+/C−) are connected between the supply voltage rails, as illustrated in example
CP0_Tsr=½.Math.Vs−Vb (330°≤θ<30°) (120)
CP1_Tsr=Va−½.Math.Vs (30°≤θ<90°) (121)
CP2_Tsr=½.Math.Vs−Vc (90°≤θ<150°) (122)
CP3_Tsr=Vb−½.Math.Vs (150°≤θ<210°) (123)
CP4_Tsr=½.Math.Vs−Va (210°≤θ<270°) (124)
CP5_Tsr=Vc−½.Math.Vs (270°≤θ<330°) (125)
(390) The commutation point is detected when the calculated CP≤0. This step is illustrated in the regenerative motor braking operation flowchart in
(391) (iii) PWM Off-Time CPD with Phase Current Equal to Zero (I=0)
(392) In this PWM off-time interval only one active phase coil (C−) is connected to the 0V rail via the C_L switch after the phase current has decayed to zero, as illustrated in the example of
CP0_Toff=½.Math.Va−Vb (330°≤θ<30°) (126)
CP1_Toff=Va−½.Math.Vb (30°≤θ<90°) (127)
CP2_Toff=½.Math.Vb−Vc (90°≤θ<150°) (128)
CP3_Toff=Vb−½.Math.Vc (150°≤θ<210°) (129)
CP4_Toff=½.Math.Vc−Va (210°≤θ<270°) (130)
CP5_Toff=Vc−½.Math.Va (270°≤θ<330°) (131)
(393) The commutation point is detected when the calculated CP≤0. This step is illustrated in the regenerative motor braking operation flowchart in
(394) (b) Synchronous Rectification During PWM Off-Time
(395) A practical example of the disclosed synchronous rectification operation during regenerative motor braking is shown in
(396) (c) Sensorless BLDCM Brake Operation Commutation and Synchronous Rectifier Phase Coils Switching States
(397) The following section outlines the phase coil switching states during PWM on-time and off-time intervals in the disclosed BEMF sensorless 60° step regenerative motor braking operation. The CP and corresponding PWM on-time and synchronous rectifier phase coil switching states for one complete electrical revolution in a BLDC motor are illustrated in
(398) (i) PWM On-Time Interval Active Phase Coils
(399) Table 20 shows the active phase coils during PWM T1 on-time interval of operation
(400) TABLE-US-00022 TABLE 20 PWM on-time active phase coils during sensorless motor brake operation Rotor Position Commutation Top Active Bottom Active Sector θ State Phase Coil Phase Coil 330-30° 0 — C−, A− 30-90° 1 — C−, B− 90-150° 2 — A−, B− 150-210° 3 — A−, C− 210-270° 4 — B−, C− 270-330° 5 — B−, A−
(401) (d) PWM Off-Time Synchronous Rectifier Interval Active Phase Coils
(402) Table 21 shows the active phase coils during PWM off-time interval of operation, employing the synchronous rectifier technique to control the phase current.
(403) TABLE-US-00023 TABLE 21 PWM off-time synchronous rectifier active phase coils during sensorless motor brake operation Rotor Position Commutation Synchronous Rectifier Bottom Active Sector θ State (Top Active) Phase Coil Phase Coil 330-30° 0 A+ C− 30-90° 1 B+ C− 90-150° 2 B+ A− 150-210° 3 C+ A− 210-270° 4 C+ B− 270-330° 5 A+ B−
(404) (e) PWM Off-Time (I=0) Active Phase Coils
(405) Table 22 shows the active phase coils during PWM off-time interval of operation when the motor biking phase currents have decayed to zero (I=0).
(406) TABLE-US-00024 TABLE 22 PWM off-time (I = 0) active phase coils during sensorless motor brake operation Rotor Position Commutation Top Active Bottom Active Sector θ State Phase Coil Phase Coil 330-30° 0 — C− 30-90° 1 — C− 90-150° 2 — A− 150-210° 3 — A− 210-270° 4 — B− 270-330° 5 — B−
(407) (f) Regenerative Motor Braking Operation Extensions
(408) The disclosed sensorless BLDC motor regenerative braking and software controlled synchronous rectification has been presented for the case of high-side PWM switching topology, commonly used with bootstrapped transistor gate drivers. However, the disclosed method can be applied to any other PWM switching topology, such as for example, low-side PWM switching and alternate low-high side PWM switching. The disclosed method can also be applied to any sensored brushless motor braking application employing rotor position detection sensors, such as Hall Effect sensors used to perform electric brushless motor commutation.
(409) Sensorless BLDCM Controller Integration
(410) The complete integration of the disclosed sensorless BLDC motor controller embodiments is shown in
(411) (a) High Speed BLDC Motor Start
(412) If a BLDC motor has high rotational speed at the start, which for example is commonly encountered in mobility equipment such as electric vehicles, then the BEMF voltage waveform magnitudes and phases can be used to deduce the initial rotor position and then consequently engage the low to high speed sensorless BLDCM controller operation, as shown in
(413) (b) Crossover Sensorless BLDCM Operation
(414) A crossover to high or zero-to-low speed sensorless BLDCM operation can be performed when a set motor speed level is reached. This method has been found to work well in most practical BLDCM applications. Combinations with other methods, for example, such as when a set PWM duty cycle level is demanded can also be implemented.
(415) (c) Sensorless BLDCM Controller Integration Extensions
(416) In certain BLDCM application where robust start-up provided by the zero to low speed SBLDCM operation is not required, it is also possible to start and operate a BLDC motor directly with the disclosed low to high speed SBLDCM controller operation, which is capable of operating motors near zero speed without rotation direction detection and with increased start-up motor torque.
(417) Three-Phase Optimized Power Control PCB Layout
(418) This section discloses the three-phase power control PCB layout design with an improved functional performance over the prior art PCB designs. It is evident that the PCB layouts used in three-phase power control applications generally consist of a rectangular power transistor layout configuration. One such example is disclosed in U.S. Pat. No. 7,154,196.
(419) In three-phase power control applications, such as professional R/C electric vehicle racing these prior art rectangular power transistor PCB layout designs as shown in
(420) 1) Increased output phase resistances and power losses due to relatively large PCB copper track distance between the external power wire connections and power transistors furthest away from the wire connections (for example PCB track distance between C4 power transistor and phase C power wire connection in
(421) 2) Increased heating of the middle phase group (B1-B4 transistor group in
(422) 3) Unbalanced phase resistances and inductances due to different PCB track distances between the power transistors and the output power wire connections (for example, A1-A4 transistor group and GND power wire connection in
(423) In many three-phase power control applications, such as R/C electric vehicle motor controllers described in [30] and [31], it is necessary to provide power wire connections outside of the controller casing to allow the user to replace the wires when required. This requirement restricts the possible placement of the power wire pads locations on the PCB with respect to the power transistors resulting in reduced PCB performance with the rectangular PCB layouts.
(424) (a) Radially Symmetrical Three-Phase Power Controller PCB Layout
(425) The disclosed three-phase power control PCB layout configuration improves on the problems of the prior art designs with a novel radially symmetrical power PCB layout. Practical examples of three different embodiments implemented in practice are shown in
(426) (i) Key Design Differences of the Radially Symmetrical Three-Phase Power Controller PCB Layout
(427) As shown in example embodiments in
(428) 1) Groups of paralleled (or single) power transistors (low-side and high-side switching) belonging to each phase are placed in radial symmetry about the center of PCB on the top and, or bottom PCB layers.
(429) 2) Phase output power wire connections are placed in the middle of each power transistor group geometry, thus physically reducing the PCB copper track resistance in each phase.
(430) (ii) Key Performance Advantages of the Radially Symmetrical Three-Phase Power Controller PCB Layout
(431) The placement of power transistors in radial symmetry with respect to the center of the PCB makes it possible to address the following challenges in three-phase power control applications:
(432) 1) Reduced phase output resistances in a similar PCB footprint size.
(433) 2) Reduced PCB power losses and thermal losses and increased efficiency.
(434) 3) More even heat distribution amongst the power transistors in each phase, reducing regions of hot spots.
(435) 4) More balanced phase resistances and inductances in all power transistor phase groups, reducing the likelihood of power transistor damage during high speed switching and high current power applications.
(436) 5) Reduced PCB manufacturing costs due to the reduced PCB copper track thickness and lesser number of layers required to achieve the same output efficiency as prior art designs.
(437) 6) Improved power PCB performance with controller case designs requiring external power wire connections to the PCB.
(438) (iii) Electrical Connections of Low-Side and High-Side Power Transistors of the Radially Symmetrical Three-Phase Power Controller PCB Layout
(439) In the disclosed radially symmetrical three-phase power PCB layout configuration, such as shown in the example embodiments in
(440) Groups consisting of only the low-side (A_L, B_L, C_L) or only the high-side (A_H, B_H, C_H) switching power transistors (paralleled or single) placed on the same PCB layer.
(441) One such practical embodiment is shown in
(442) Groups consisting of both the low-side (A_L, B_L, C_L) and high-side (A_H, B_H, C_H) switching power transistors (paralleled or single) placed on the same PCB layer.
(443) One such practical embodiment is shown in
(444) In the embodiment shown in
(445) (b) Practical Embodiment of Radially Symmetrical Three-Phase Power Controller PCB Layout
(446) One practical embodiment of the disclosed radially symmetrical three-phase optimized PCB layout, implemented in R/C electric vehicle BLDCM control applications, having four paralleled power transistors in each phase group and power wire connections placed outside the controller casing is shown in
(447) Top layer consisting of paralleled low-side switching power transistors A1-A4 (A_L), B1-B4 (B_L) and C1-C4 (C_L); Bottom layer consisting of paralleled high-side switching power transistors A5-A8 (A_H), B5-B8 (B_H) and C5-C8 (C_H); Connector (CN1) provides electrical connections to adjoining PCB containing embedded microcontroller and low power signal circuitry; Internal GND power copper plane layer; Internal Vs power copper plane layer.
(448) The internal GND and Vs power copper plane layers provide a low resistance connection between the groups of power transistors in each phase (A, B, C) and the GND and Vs external power wire connection pads. In practice each power plane layer (GND, Vs) consists of two or more internal copper layers of the same design stacked on top of each other within the PCB to reduce the power connection resistance as low as possible. The connections between the top, bottom and internal layers are provided by the small diameter through-hole vias. Thus a complete power PCB design can include a total of six or more copper layers.
(449) (c) Phase Circuit Resistance Analysis
(450) The following section presents an analysis and comparison of the output phase resistances for the prior art rectangular PCB layout and the disclosed radially symmetrical PCB layout.
(451) (i) Rectangular PCB Layout Phase Circuit Resistance Analysis
(452)
Rph_rec=1/(1/(2.Math.R+1/(1/(2.Math.R+1/(1/Rt+1/(2.Math.R+Rt)))+1/Rt))+1/Rt)+2.Math.R (132)
(453) (ii) Radially Symmetrical PCB Layout Phase Circuit Resistance Analysis
(454)
Rph_sym=½.Math.(2.Math.R+1/(1/Rt+1/(2.Math.R+Rt))) (133)
(455) (iii) Total Phase Resistance Comparison
(456)
(457) In practical three-phase power control power PCB layout embodiments such as in
R=ρ.Math.L/(W.Math.T) (134)
(458) Where: ρ=1.69×10.sup.−8 Ω/m (copper resistivity at 20° C.); L=0.005 m (unit PCB copper length between adjacent power transistors, approximately equal to power transistor width [29]); W=0.005 m (PCB copper track width); T=100×10.sup.−6 m=0.5×10.sup.−6 m.sup.2 (PCB copper thickness 100 um (3 oz.)
(459) Thus, the unit PCB copper track resistance (R) is approximately equal to:
R=1.69×10.sup.−8 Ω/m.Math.0.005 m/(0.005 m.Math.100×10−6 m)=0.169 mΩ (135)
(460) A modern power transistor switching device such as the International Rectifiers IRFH5300PbF [29] commonly used in BLDCM power control applications has a typical RDS on-resistance (Rt) equal to 1.4 mΩ. Thus, in this example the normalized power transistor resistance (Rt) is equal to:
Rt=1.4 mΩ/0.169 mΩ≈8.3 (136)
(461) With Rt=8.3, the rectangular PCB layout has approximately 5.4/3.3≈1.6 times higher phase resistance than the radially symmetrical PCB layout, as shown in
(462) (d) Radially Symmetrical PCB Layout Extensions
(463) The disclosed radially symmetrical PCB layout can be employed in any three-phase power control applications such as, but not limited to: 1) Three-phase BLDC motor controllers (sensored, sensorless); 2) Three-phase BLDC motors with controllers built inside the motor, such as described in [32] (for example using circular PCB layout embodiments disclosed in
(464) The disclosed radially symmetrical PCB layout can be realized with any number of external and internal PCB conductive layers, for example, with a single sided PCB, a double sided PCB and a multiple layer PCB.
(465) Furthermore, the disclosed radially symmetrical PCB layout can be realized with any number of paralleled power transistor devices in each phase and with any of the available power transistor device packages and footprints such as, but not limited to: PQFN, QFN, DPAK, D2PAK, PPAK, SO8, SOP8, SOT143, SOT23, SOT223, SOT523, SOT666, SOT89, TSOP6, TSSOP8, DirectFET, FlipFET, MicroFhT, Micro8, PolarPAK, PowerPAK. Power 33, Power 56.
(466) The disclosed radially symmetrical method of power transistor and power wire placement on a PCB can also be employed in applications other than three-phase power control applications, for example, such as two-phase, four-phase or five-phase systems. In this case the radial angle between each phase is adjusted according to the number of phases, for example, 180° for two-phase, 90° for four-phase and 72° for five-phase system.
(467) Interpretation
(468) Reference throughout this specification to “one embodiment”, “some embodiments” or “an embodiment” means that a particular feature, structure or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, appearances of the phrases “in one embodiment”, “in some embodiments” or “in an embodiment” in various places throughout this specification are not necessarily all referring to the same embodiment, but may. Furthermore, the particular features, structures or characteristics may be combined in any suitable manner, as would be apparent to one of ordinary skill in the art from this disclosure, in one or more embodiments.
(469) As used herein, unless otherwise specified the use of the ordinal adjectives “first”, “second”, “third”, etc., to describe a common object, merely indicate that different instances of like objects are being referred to, and are not intended to imply that the objects so described must be in a given sequence, either temporally, spatially, in ranking, or in any other manner.
(470) In the claims below and the description herein, any one of the terms comprising, comprised of or which comprises is an open term that means including at least the elements/features that follow, but not excluding others. Thus, the term comprising, when used in the claims, should not be interpreted as being limitative to the means or elements or steps listed thereafter. For example, the scope of the expression a device comprising A and B should not be limited to devices consisting only of elements A and B. Any one of the terms including or which includes or that includes as used herein is also an open term that also means including at least the elements/features that follow the term, but not excluding others. Thus, including is synonymous with and means comprising.
(471) As used herein, the term “exemplary” is used in the sense of providing examples, as opposed to indicating quality. That is, an “exemplary embodiment” is an embodiment provided as an example, as opposed to necessarily being an embodiment of exemplary quality.
(472) It should be appreciated that in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure and aiding in the understanding of one or more of the various inventive aspects. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the Detailed Description are hereby expressly incorporated into this Detailed Description, with each claim standing on its own as a separate embodiment of this invention.
(473) Furthermore, while some embodiments described herein include some but not other features included in other embodiments, combinations of features of different embodiments are meant to be within the scope of the invention, and form different embodiments, as would be understood by those skilled in the art. For example, in the following claims, any of the claimed embodiments can be used in any combination.
(474) Furthermore, some of the embodiments are described herein as a method or combination of elements of a method that can be implemented by a processor of a computer system or by other means of carrying out the function. Thus, a processor with the necessary instructions for carrying out such a method or element of a method forms a means for carrying out the method or element of a method. Furthermore, an element described herein of an apparatus embodiment is an example of a means for carrying out the function performed by the element for the purpose of carrying out the invention.
(475) In the description provided herein, numerous specific details are set forth. However, it is understood that embodiments of the invention may be practiced without these specific details. In other instances, well-known methods, structures and techniques have not been shown in detail in order not to obscure an understanding of this description
(476) Similarly, it is to be noticed that the term coupled, when used in the claims, should not be interpreted as being limited to direct connections only. The terms “coupled” and “connected,” along with their derivatives, may be used. It should be understood that these terms are not intended as synonyms for each other. Thus, the scope of the expression a device A coupled to a device B should not be limited to devices or systems wherein an output of device A is directly connected to an input of device B. It means that there exists a path between an output of A and an input of B which may be a path including other devices or means. “Coupled” may mean that two or more elements are either in direct physical or electrical contact, or that two or more elements are not in direct contact with each other but yet still co-operate or interact with each other.
(477) Thus, while there has been described what are believed to be the preferred embodiments of the invention, those skilled in the art will recognize that other and further modifications may be made thereto without departing from the spirit of the invention, and it is intended to claim all such changes and modifications as falling within the scope of the invention. For example, any formulas given above are merely representative of procedures that may be used. Functionality may be added or deleted from the block diagrams and operations may be interchanged among functional blocks. Steps may be added or deleted to methods described within the scope of the present invention.