1. Department of Electrical Engineering Southern Taiwan University of Science and Technology Robot and Servo Drive Lab. 2016/7/8 First-Pulse Technique for Brushless DC Motor Standstill Position Detection Based on Iron B-H Hysteresis IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, VOL. 59, NO. 5, MAY 2012 P.2319~P.2328 Omar Scaglione, Miroslav Markovic, and Yves Perriard, Senior Member, IEEE 學生 : 洪瑞志 指導教授 : 王明賢

2. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 2 outline Abstract Introduction First-Pulse Standstill Technique Position Detection Objective First-Pulse Technique Bases First-Pulse Technique Theory First-Pulse Technique Simplified Model First-Pulse Technique Measurements Practical Implementation Issues Conclusion And Future Work Reference

3. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 3 Abstract This paper introduces a promising technique, called the first- pulse technique, for detecting the rotor position of a brushless DC motor at standstill. The technique is based on the B-H hysteresis characteristics of the stator iron. In the first-pulse technique, the rotor position is estimated not by an absolute measurement, but by the comparison between the measurated effects created by the first injected pulse and the following ones. This technique permits to detect the electrical position of a non-salient rotor.

4. Department of Electrical Engineering Southern Taiwan University of Science and Technology Introduction In fact, the position detection at zero speed after a hard reset of the driving system (when the position estimation is lost) is very difficult to achieve if no saliencies are present on the rotor, like in [19], [20]. Many authors have also modified the rotor geometries for improving the sensorless control of PM motor without affecting the torque [21]–[29]. Very few techniques, as for example [30] where the magnetic anisotropy of the permanent magnet (PM) is sensed, or techniques based on saturation [31]–[33], work for a nonsalient BLDC motor. In this paper, a new and innovative technique, called first-pulse technique, able to detect the rotor position of a mechanically non-salient motor at standstill is introduced. 2016/7/8 Robot and Servo Drive Lab. 4

5. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 5 First-Pulse Standstill Technique Position Detection Objective Fig. 1. Hall sensor states for a complete electrical period. Shows the Hall sensor states of a complete electrical period of the test BLDC motor used in this paper. Acronyms.

6. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 6 First-Pulse Technique Bases Fig. 2. Analyzed motor configuration. A small industrial BLDC motor (100W, 3 phases, 10 poles, Un = 60 V, In = 6.2 A) with mechanically non-salient rotor is analyzed. ( Rated voltage, rated current.) A. Physical Bases.

7. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 7 First-Pulse Technique Bases Fig. 3. Stator tooth iron B-H characteristic [34]. It shows the B-H hysteresis of the stator iron. The external loop, the bigger hysteresis, is obtained when the rotor moves in front a stator tooth. If an AC signal is injected in a phase, the tooth iron working point moves along the B-H characteristic following a local loop. In Fig. 3, several local loops, corresponding to several rotor positions, are plotted. A. Physical Bases. External loop Local loops

8. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 8 First-Pulse Technique Bases Fig. 4. Proposed measurement circuit [34]. The circuit shown in Fig. 4 is used for detecting the motor phase impedance variations. The motor is schematically represented in Fig. 4 by its three-phase impedances, Z1, Z2, Z3, and the neutral point N. External resistances R are added to create an artificial neutral point N′. (N Star-connected motor neutral point. N′ Star-connected motor artificial neutral point. Z1, Z2, Z3 Three-phase motor impedances.) B. Measurement Circuit.

9. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 9 First-Pulse Technique Bases B. Measurement Circuit. The analytical solution of in this circuit is This expression varies with R. Equation (1) can be simplified if R Zi, i.e., R→∞. The calculation of this limit applied to (1) gives Thus, for sufficiently high values of R, just Z1 and Z2 influence and the R term disappears. ( Voltage appearing when is applied.)

10. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 10 First-Pulse Technique Bases Fig. 5. PWM pulse of, and corresponding and. B. Measurement Circuit. The PWM signal driving the voltage with = 20 μs, the current and voltage after filtering and amplification, as used for the measurements done in this paper.

11. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 11 First-Pulse Technique Theory Fig. 6. Iron BH hysteresis loops, NSN rotor position. 1st pulse [35]. The first-pulse technique consists in the injection of two succesive pulses of and the measurement of. An important fact is that the position information is not retrieved by the absolute value of as in [34], [40], but by the comparison between the signals created by the first and the second pulse. NSN and NNS. For simplifying the phenomenon description. A. Position NSN Physical Interpretation. The in /decreasing current.

12. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 12 First-Pulse Technique Theory Fig. 7. Iron BH hysteresis loops, NSN rotor position. 2nd pulse [35]. A. Position NSN Physical Interpretation. It can therefore be stated that there is a magnetic difference in the stator iron between the first pulse and all the following ones for this position. This difference influences the phase impedance, and can therefore be measured in the voltage.

13. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 13 First-Pulse Technique Theory Fig. 8. Iron BH hysteresis loops, NNS rotor position. 1st pulse [35]. B. Position NNS Physical Interpretation. Contrary to the position NSN, in position NNS. As a consequence, every pulse of applied in this position will have the same effect on the stator iron and no difference in the magnetic path will appear between the first pulse and the following ones. Hence, no difference will be measured between the first signal and the following ones.

14. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 14 First-Pulse Technique Simplified Model Fig. 9. Iron B-H hysteresis loop, chosen angular reference. In this paper, it is chosen to fix the angular position origin (0 ◦ ) at the beginning of the iron North polarization.

15. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 15 First-Pulse Technique Simplified Model Fig. 10. Iron B-H hysteresis loop, chosen angular origin. It is chosen to place the motor phase 2 at 0 °, phase 1 at 120 ° and phase 3 at 240 ° on the same B-H hysteresis. A. Phases Iron B-H Working Point.

16. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 16 First-Pulse Technique Simplified Model Fig. 11. Iron B-H hysteresis loop, other possible angular origin. It is shown how, for the same rotor position of 0° (NNS), the phases 1 and 2 can have another location on the B-H hysteresis. A. Phases Iron B-H Working Point.

17. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 17 First-Pulse Technique Simplified Model TABLE I THEORICAL RESULTS ON THE SINGLE PHASES Zi EVERY 15° B. Prediction of the First-Pulse Effect on the Phase Impedances. Table I and all the measurements and examples in this paper are based on the angular origin shown in Fig. 10. Starting from this angular reference, the working point of the three phases is automatically rotated counterclockwise every 15° on the B-H loop.

18. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 18 First-Pulse Technique Simplified Model B. Prediction of the First-Pulse Effect on the Phase Impedances. it can be seen how the positive current in the phase 1 creates the first-pulse effect on Z1, and how the negative current in the phase 2 creates the effect too, on Z2. It can therefore be easily predicted that for this position pulse 12 creates the first pulse effect on. Fig. 12. Iron B-H hysteresis loop, theorical effect prediction of pulse 12 on based on the simplified model. Position 195 °.

19. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 19 First-Pulse Technique Simplified Model C. Prediction of the First-Pulse Effect on. It can furthermore be seen from (2) that the contributions of two phases are merged in the voltage for every possible pulse. For each angular position and for each of the 6 possible pulses that can be imposed to a star-connected BLDC motor.

20. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 20 First-Pulse Technique Simplified Model D. Interpretation. The first remark to point out on the theorical results presented in Table II is that the phenomenon does not have the same borders as the Hall states. This is rapidly understood from the observation of Fig. 9 in which clearly appears that the phenomenon is related to the phase position on the B-H hysteresis loop. With the first-pulse technique, the rotor position can be identified within a range of 45° that overlaps two 60° Hall states. A more accurate position estimation is not achievable without an additional information, but this is enough for starting up the motor.

21. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 21 First-Pulse Technique Measurements A. Measurements Setup Fig. 13. Measurement setup. Measurements are obtained as follows: three pulses, like the one shown in Fig. 5, are injected by a custom electronics into the motor phases. Between every pulse a =130 μs is waited, so that goes back to its steady state before the next pulse. Signals,, and of the three pulses are measured with a scope and then saved on a PC.

22. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 22 First-Pulse Technique Measurements A. Measurements Setup. Fig. 14. 1st row: original signals. 2nd row: shifted signals. 3rd row: 1st and 2nd rows superimposed.

23. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 23 First-Pulse Technique Measurements B. Measurements. Fig. 15. Hall sensor states for a complete electrical period. Measurements are performed for many positions on a complete rotor electrical revolution, maintaining the same signal, pulse 12, for all the trials. Fig. 15 shows the electrical angular position of the performed measurements.

24. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 24 First-Pulse Technique Measurements B. Measurements. Fig. 16. Impulsion 12, position 18 °. Theory (left), measurement (right). For the position 18° (NNS).

25. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 25 First-Pulse Technique Measurements B. Measurements. Fig. 17. Impulsion 12, position 108 °. Theory (left), measurement (right). For the position 108° (SNS).

26. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 26 First-Pulse Technique Measurements B. Measurements. Fig. 18. Impulsion 12, position 234 °. Theory (left), measurement (right). For the position 234° (SSN).

27. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 27 First-Pulse Technique Measurements C. Interpretation. Practical Implementation Issues

28. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 28 Conclusion And Future Work A. Conclusion. The introduced first-pulse position detection technique is very promising, as almost every PM motor has an iron stator. Morevover, it has to be pointed out that the technique does not need any previous calibration because the position is estimated by the comparison between two measurements, and not using an absolute value. In this case, position resolution is 45°. B. Future Work. Because of the particular physical origin of the phenomenon, simulations of the first-pulse effect are not possible with the actual finite elements softwares. The introduced theory is still very empirical and surely needs to be refined. Future work may therefore first of all focus on the physical theory for improving the understanding of the phenomenon and refinement of the measurement predictions. Finally, high priority should be placed on a way for resetting the stator iron to a known position on the B-H characteristic, in order to ensure repeatability of the measurements and independence from the past.

29. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 29 References [1] F. Genduso, R. Miceli, C. Rando, and G. Galluzzo, “Back EMF sensorless-control algorithm for high-dynamic performance PMSM,” IEEE Trans. Ind. Electron., vol. 57, no. 6, pp. 2092–2100, Jun. 2010. [2] C. Bianchini, C. Concari, and A. Toscani, “Low-cost sensorless BLDC for organic fluids treatment in sterile environments,” in Proc. 18th ICEM, Sep. 2008, pp. 1–5. [3] S. Bujacz, A. Cichowski, P. Szczepankowski, and J. Nieznanski, “Sensorless control of high speed permanent-magnet synchronous motor,” in Proc. 18th ICEM, Sep. 2008, pp. 1–5. [4] T. Ichikawa and H. Dohmeki, “Field expansion of low speed in sensorless 120-degree conduction drives of brushless DC motor,” in Proc. XIX ICEM, Sep. 2010, pp. 1–6. [5] G. Foo and M. Rahman, “Sensorless direct torque and flux-controlled ipm synchronous motor drive at very low speed without signal injection,” IEEE Trans. Ind. Electron., vol. 57, no. 1, pp. 395–403, Jan. 2010. [6] K. Hanamura and H. Dohmeki, “Position sensorless control for interior permanent magnet synchronous motor using adaptive flux observer,” in Proc. XIX ICEM, Sep. 2010, pp. 1–6. [7] G. Foo and M. Rahman, “Sensorless sliding-mode MTPA control of an IPM synchronous motor drive using a sliding-mode observer and HF signal injection,” IEEE Trans. Ind. Electron., vol. 57, no. 4, pp. 1270– 1278, Apr. 2010.

30. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 30 References [8] X. Fei, W. Hao-Xiong, C. Min-Liang, and L. Yong, “A novel sensorless control of PMSG based on sliding mode observer,” in Proc. XIX ICEM, Sep. 2010, pp. 1–4. [9] S. Sayeef, G. Foo, and M. Rahman, “Very low speed operation of a variable structure direct torque controlled IPM synchronous motor drive using combined hf signal injection and sliding observer,” in Proc. 18th ICEM, Sep. 2008, pp. 1–6. [10] S. Sayeef, G. Foo, and M. Rahman, “Rotor position and speed estimation of a variable structure direct-torque-controlled IPM synchronous motor drive at very low speeds including standstill,” IEEE Trans. Ind. Electron., vol. 57, no. 11, pp. 3715–3723, Nov. 2010. [11] G. Scelba, G. De Donato, F. Capponi, A. Consoli, and O. Honorati, “A co-simulation platform for evaluation of sensorless control techniques for IPMSMS,” in Proc. XIX ICEM, Sep. 2010, pp. 1–7. [12] S. Bolognani, S. Calligaro, R. Petrella, andM. Tursini, “Sensorless control of IPM motors in the low-speed range and at standstill by HF injection and DFT processing,” IEEE Trans. Ind. Appl., vol. 47, no. 1, pp. 96–104, Jan./Feb. 2011. [13] N. Bianchi, S. Bolognani, J.-H. Jang, and S.-K. Sul, “Comparison of PM motor structures and sensorless control techniques for zero-speed rotor position detection,” IEEE Trans. Power Electron., vol. 22, no. 6, pp. 2466– 2475, Nov. 2007.

31. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 31 References [14] N. Bianchi and S. Bolognani, “Influence of rotor geometry of an IPM motor on sensorless control feasibility,” IEEE Trans. Ind. Appl., vol. 43, no. 1, pp. 87–96, Jan./Feb. 2007. [15] N. Bianchi, S. Bolognani, J.-H. Jang, and S.-K. Sul, “Advantages of inset PM machines for zero-speed sensorless position detection,” IEEE Trans. Ind. Appl., vol. 44, no. 4, pp. 1190–1198, Jul./Aug. 2008. [16] R. Raute, C. Caruana, C. Staines, J. Cilia, M. Sumner, and G. Asher, “Analysis and compensation of inverter nonlinearity effect on a sensorless PMSM drive at very low and zero speed operation,” IEEE Trans. Ind. Electron., vol. 57, no. 12, pp. 4065–4074, Dec. 2010. [17] O. Benjak and D. Gerling, “Review of position estimation methods for IPMSM drives without a position sensor Part I: Nonadaptive methods,” in Proc. XIX ICEM, Sep. 2010, pp. 1–6. [18] O. Benjak and D. Gerling, “Review of position estimation methods for IPMSM drives without a position sensor Part II: Adaptive methods,” in Proc. XIX ICEM, Sep. 2010, pp. 1–6. [19] T. Ohnuma, S. Doki, and S. Okuma, “Extended EMF observer for wide speed range sensorless control of salient-pole synchronous motor drives,” in Proc. XIX ICEM, Sep. 2010, pp. 1–6. [20] F. De Belie, T. Vyncke, and J. Melkebeek, “Parameterless rotor position estimation in a direct-torque controlled salient-pole PMSM without using additional test signals,” in Proc. XIX ICEM, Sep. 2010, pp. 1–6.

32. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 32 References [21] M. Tomita, M. Satoh, H. Yamaguchi, S. Doki, and S. Okuma, “Sensorless estimation of rotor position of cylindrical brushless DC motors using Eddy current,” in Conf. Rec. 4th Int. Workshop AMC-MIE, 1996, vol. 1, pp. 24–28. [22] M. Tomita, S. Doki, S. Okuma, and H. Yamaguchi, “Sensorless rotor position estimation at standstill of cylindrical brushless DC motors using opened phase voltage change caused by Eddy currents,” Elect. Eng. Jpn., vol. 126, no. 1, pp. 52–60, Jan. 1999. [23] N. Bianchi and S. Bolognani, “Sensorless-oriented design of PM motors,” IEEE Trans. Ind. Appl., vol. 45, no. 4, pp. 1249–1257, Jul./Aug. 2009. [24] A. Faggion, S. Bolognani, and N. Bianchi, “Ringed-pole permanent magnet synchronous motor for position sensorless drives,” in Proc. ECCE, 2009, pp. 3837–3844. [25] N. Bianchi, S. Bolognani, and A. Faggion, “Rotor design arrangement of SPM motors for the sensorless control at low speed and standstill,” in Proc. 14th Int. EPE/PEMC, 2010, pp. S1-23–S1-28. [26] Y. Kano, T. Kosaka, N. Matsui, and T. Nakanishi, “Sensorless-oriented design of IPM motors for general industrial applications,” in Proc. 18th ICEM, Sep. 2008, pp. 1–6. [27] Y. Kano, T. Kosaka, N.Matsui, and T. Nakanishi, “Design and experimental verification of a sensorless-oriented concentrated-winding IPMSM,” in Proc. XIX ICEM, Sep. 2010, pp. 1–6.

33. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 33 References [28] P. Sergeant, F. De Belie, and J. Melkebeek, “Rotor geometry design of an interior permanent-magnet synchronous machine for more accurate sensorless control,” in Proc. XIX ICEM, Sep. 2010, pp. 1–6. [29] L. Alberti, N. Bianchi, S. Bolognani, and E. Fornasiero, “IM rotor parameters analysis with an intentionally created saliency,” in Proc. 1st Symp. SLED, 2010, pp. 120–126. [30] J. Persson, M. Markovic, and Y. Perriard, “A new standstill position detection technique for nonsalient permanent-magnet synchronous motors using the magnetic anisotropy method,” IEEE Trans.Magn., vol. 43, no. 2, pp. 554–560, Feb. 2007. [31] M. Schroedl, “Detection of the rotor position of a permanent magnet synchronous machine at standstill,” in Proc. ICEM, Pisa, Italy, Sep. 1988. [32] F. Demmelmayr, A. Eilenberger, and M. Schroedl, “Sensorless electric traction drive with 500 nm outer rotor permanent magnet synchronous machine,” in Proc. XIX ICEM, Sep. 2010, pp. 1–7. [33] L. Cardoletti, A. Cassat, and M. Jufer, “Indirect position detection at standstill for brushless DC and step motors,” in Proc. EPE 3rd Eur. Conf. Power Electron. Appl., Aachen, Germany, Oct. 9–12, 1989, vol. 3, pp. 1219–1222. [34] O. Scaglione, M. Markovic, and Y. Perriard, “Exploitation of iron B-H local hysteresis for the rotor position detection of a PM motor,” in Proc. IEMDC, Miami, FL, 2009, pp. 1641–1646.

34. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 34 References [35] O. Scaglione, M. Markovic, and Y. Perriard, “Exploitation of a new iron B-H phenomenon for the standstill position detection of PM motors,” in Proc. ICEM, Rome, Italy, 2010, pp. 1–4. [36] G. Bertotti, Hysteresys in Magnetism—For Physicists, Material Scientists and Engineers. New York: Academic, 1998. [37] T. Sato and Y. Sakaki, “Physical meaning of equivalent loss resistance of magnetic cores,” IEEE Trans. Magn., vol. 26, no. 5, pp. 2894–2897, Sep. 1990. [38] V. Basso and G. Bertotti, “Hysteresis models for the description of domain wall motion,” IEEE Trans. Magn., vol. 32, no. 5, pp. 4210–4212, Sep. 1996. [39] J. Shao, D. Nolan, and T. Hopkins, “A novel direct back EMF detection for sensorless brushless DC (BLDC) motor drives,” in Proc. IEEE APEC, Mar. 2002, pp. 33–37. [40] O. Scaglione, M. Markovic, and Y. Perriard, “PM motor sensorless position detection based on iron B-H local hysteresis,” in Proc. ICEMS, Tokyo, Japan, 2009, pp. 1–6. [41] Keithley, Low Level Meausurements Handbook. Keithley, Cleveland, OH, 2009. [Online]. Available: http://www.keithley.com

35. Department of Electrical Engineering Southern Taiwan University of Science and Technology 2016/7/8 Robot and Servo Drive Lab. 35 ~The End~ Thanks