E XPERIMENTAL R ESULTS OF THE C OMMUTATION C ONTROL

The sensorless control method is realized in software. Fig.6.1 and Fig.6.2 will be show the block diagram in Matlab^®−Simulink^®. Fig.6.1 is the overall sensorless control algorithm with three terminal voltage inputs and three estimated Hall-sensor output produced from the algorithm. Fig.6.2 is the block diagram frequency independent phase shifter (FIPS).

Fig.6.1 The block diagram of overall sensorless control algorithm

Tp Cp

The three terminal voltages measured directly from the BLDC motor are shown in Fig.6.3 and 6.4. To analyze the experimental results at high angular velocity (about 170.5 rpm) and low angular velocity (about 49.7 rpm), it could be found that the zero crossing point in a low angular velocity is hard to detect. Therefore, an additional low-pass filter is needed before the voltages are transferred into the PC. By analyzing the frequency spectrum as shown in Fig.6.5, a set of low-pass filter could be designed to eliminate the unwanted noises as described in Section 4.4. After using the low-pass filter, the frequency spectrum of terminal voltage is shown in Fig.6.6. Fig.6.7 shows the terminal voltage in a low angular velocity after using the low-pass filter.

Fig.6.2 The block diagram of frequency independent phase shifter (FIPS)

Cn Tn

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5

Fig.6.3 Terminal voltage at high angular velocity

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 V b(volt)V c(volt)

Fig.6.4 Terminal voltage at low angular velocity

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5x 10⁴

Radian Frequency (ω)

Amplitude

π ²π

0

Fig.6.5 Frequency spectrum of terminal voltage at low angular velocity

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5x 10⁴

Radian Frequency(ω)

Amplitude

π 2π

0

Fig.6.6 Frequency spectrum of terminal voltage after using the low-pass filter

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0

5 10 15

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0

5 10 15

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0

5 10 15

time(sec) V c(volt)V b(volt)V a(volt)

Fig.6.7 Terminal voltage in a low angular velocity after using the low-pass filter

After using the low-pass filter, the zero-crossing points are easier to detect for sensorless algorithm. Then the experimental results of the sensorless control in high angular velocity will be shown in Fig.6.8 and Fig.6.9; and a low angular velocity will be shown in Fig.6.10 and Fig.6.11. Finally, the comparisons between the sensorless produced signals and conventional Hall-effect signals will be shown in Fig.6.12 and Fig.6.13.

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5

Fig.6.8 Sensorless control performance at high angular velocity (I)

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5

Fig.6.9 Sensorless control performance at high angular velocity (II)

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5

Fig.6.10 Sensorless control performance at low angular velocity (I)

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5

Fig.6.11 Sensorless control performance at low angular velocity (II)

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5

Fig.6.12 Comparison of the sensorless signals and conventional Hall-effect signals at high angular velocity

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5

Fig.6.13 Comparison of the sensorless signals and conventional Hall-effect signals at low angular velocity

6.2 Experimental results of the start-up procedure

The start-up algorithm for initial position detection has been verified in Chapter 4.

Fig.6.14 shows the six current responses when the rotor is at standstill in a certain position. A voltage pulse is applied for all six segments, respectively, of an electrical cycle during 500μs, to detect the stationary position of a rotor. The current is measured by three current sensors LP-25P, which is directly getting the current information from the motor. However, the residue in the magnetic field would affect the measured current responses. Thus, the longer distance of each pulse or import the inverse current will improve the results. On the other hand, the current noise which is chattering between is clearly to see when the current is zero due to the electromagnetic interface, connector plugs and conducting wires. But the noises would not affect the detecting result since the detected current responses are large enough.

Besides, based on the polarity of ddi in Table 4.2 and Fig.4.5, the relative position of a rotor is found to be between 0° to 60° with sector 0 (see Table 4.3).

1

± 0.

Then set up the command signal with a counter as shown in Fig.6.15. In addition, take the initial sector as the initial value in the counter, and the period of pulse generator will depend on the desired angular velocity when the motor is start-up. The period can refer to Table 4.4. Fig.6.16 shows the modified open-loop start-up form sector 0. Since the torque of the BLDC motor is very high and the back-EMFs are hard

to detect at very low angular velocity, there exists the jump position at starting in the Back-EMF detected sector. But the right direction rotating is produced. Fig.6.17 shows the difference between command sector and Back-EMF detected sector. This figure could be represented that the Back-EMF could be detected after about 2 second because the difference sector position is small than 1.

0 1 2 3 4 5 6

0 0.5 1 1.5 2 2.5 3 3.5 4

current

i1+ i

1- i

2+

i2- i

3+ i₃

-time

Fig.6.14 Measured the six current responses of an stationary rotor

Fig.6.15 The block diagram of start-up procedure

0 0.5 1 1.5 2 2.5 3 0

1 2 3 4 5

time(sec)

Se c to r

0 0.5 1 1.5 2 2.5 3

0 1 2 3 4 5

time(sec)

B a ck- E M F det e c ted sect or

Hall sensor command

Fig.6.16 The modified open-loop start-up form sector 0

Fig.6.17 The difference between command sector and Back-EMF detected sector

Besides, the switch condition from the open-loop start-up mode to the sensorless commutation mode will be presented as follows. Fig.6.18 will show the block diagram of the whole sensorless system and the switch condition signal with 0 for start-up mode

and 1 for commutation mode. In Fig.6.18 (b), the counter is set for detecting the

diff_on signal.Once the difference sector is larger than 2 and smaller than 4 during the

0.5 sec, this counter will be reset and the switch signal is still 0 until another period which the diff_on signal maintained 0 for 0.5 sec.

Fig.6.19 and Fig.6.20 will show the experimental results of the motor from standstill to commutation mode at low angular velocity. Fig.6.21 and Fig.6.22 will show the results from standstill to commutation mode at high angular velocity.

Fig.6.18 (a) The block diagram of the whole sensorless system and (b) the switch condition signal

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

open-loop commutation mode

Fig.6.19 Results from standstill to commutation mode at low angular velocity (I)

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

Diff-on signal

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

open-loop commutation mode

Fig.6.20 Results from standstill to commutation mode at low angular velocity (II)

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

open-loop mode commutation mode system start

Fig.6.21 Results from standstill to commutation mode at high angular velocity (I)

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5

open-loop mode commutation mode

Fig.6.22 Results from standstill to commutation mode at high angular velocity (II)

Chapter 7 Conclusions

This thesis has analyzed and then realized a sensorless drive for the BLDC motor.

The proposed sensorless control algorithm is implemented in a real-time motor drive, and is shown successful through the experimental results.

To identify the initial rotor position of a BLDC motor, a method is presented based on the inductance variation. Once the initial rotor position is attained, the motor can be driven from standstill effectively by a modified open-loop method even though the back-EMFs could not be exactly detected owing to the noises near zero-crossing point. After the BLDC motor starts, it is required to detect the commutation instants in phase, which is implemented by a back-EMF based position estimated method. This position estimated method adopts the conventional frequency-independent phase shifter and an additional low-pass filter designed for detecting the zero-crossing points at low angular velocity.

On the other hand, in order to improve the performance, the implementation of hardware using DSP card is worth proceeding since the sampling rate can reach 120 MIPS. In this way, the current will be measured more precisely such that the initial position will be detected more accurately. Besides, the sensorless control at a very low speed is not suitable to use the back-EMF based position estimated method because the

current is discontinuous in a low angular velocity 10 rpm. Therefore, the discontinuity should be taken into account to improve the performance while current model based sensorless approaches are adopted.

In summary, the experimental results verify that the sensorless strategy is a useful alternative for applications that require high efficiency, high performance, and robust BLDC motor drives.

Reference

[1] K. Y. Cheng and Y. Y. Tzou, “Design of a Sensorless Commutation IC for BLDC Motors,” IEEE Transactions on Power Electronics, vol.18, no.6, pp.1365-1375, November 2003.

[2] J. Shao, D. Nolan, and T. Hopkins, “Improved direct back EMF detection for sensorless brushless DC (BLDC) motor drives,” in Proc. IEEE-APEC Conf., 2003, pp.300-305.

[3] D. H. Jung and I. J. Ha, “Low-cost sensorless control of brushless DC motors using frequency-independent phase shifter,” IEEE Transactions on Power Electronics, vol.15, no.4, pp.744-752, July 2000.

[4] B. Terzic and M. Mohan, “Design and Implementation of the Extended Kalman

Filter for the Speed and Rotor Position Estimation of Brushless DC Motor,” IEEE

Transactions on Industrial Electronics, vol.48, no.6, pp. 1065-1073, December

2001.

[5] J. C. Moreira, “Indirect Sensing for Rotor Flux Position of Permanent Magnet AC Motors Operating Over a Wide Speed Range,” IEEE Transactions on Industry

Applications, vol.32, no.6, pp.1394-1401, November 1996.

[6] S. Ogasawara and H. Akagi, “An approach to position sensorless drive for brushless dc motors,” IEEE Tran. Industry Applications, vol. 27, no. 5, pp.

928-933, Sep. /Oct. 1991.

[7] S. J. Kang and S. K. Sul, “Direct torque control of brushless DC motor with nonideal trapezoidal back EMF,” IEEE Trans. Power Electron., vol.10, no. 6, pp.796-802, Nov. 1995.

[8] P. Pillay and R. Krishnan, “Modeling, simulation, and analysis of

permanent-magnet motor drives, part II: The brushless dc motor drive,” IEEE

Trans. Industry Applications, vol. 25 no. 2 pp.274-279 Mar./Apr. 1989.

[9] P. Pillay and R. Krishnan, “Modeling, simulation, and analysis of

permanent-magnet motor drives, part I: The permanent magnet synchronous motor drive,” IEEE Trans. Industry Applications, vol. 25 no.2 pp.265-273 Mar.

/Apr. 1989.

[10] G. H. Jang, J. H. Park and J. H. Chang, “Position detection and start-up algorithm

of a rotor in a sensorless BLDC motor utilizing inductance variation,” IEE

Proc.-Electr. Power Applicat., vol. 149, no. 2, pp.137-142, March 2002.

[11] M. Tursini, R. Petrella, and F. Parasiliti, “Initial Rotor Position Estimation Method for PM Motors,” IEEE Tran. Industry Application, vol.39, no.6, pp.1630-1640, Nov. /Dec. 2003.

[12] J. P. Johnson, M. Ehsani, and Y. Guzelgunler, “Review of Sensorless Methods for

Brushless DC,” IEEE Trans. Industry Applications Conference, Thirty-Fourth IAS

Annual Meeting, vol. 1, pp.143-150, 3-7 October 1999.

[13] T. Yoon, “Stator design consideration of a Brushless DC motor for robust rotor position detection in inductive sense start-up,” IEEE Trans. Magnetic, vol. 42, no.

3, March 2006.

[14] IR2110/2113(S) Data Sheet No.PD60147-O from International Rectifier

Company.

[15] PCI-6024E Data Sheet from National Instruments.

[16] IRF640N, HEXFET^® Power MOSFET, Data Sheet No.PD94006 from

International Rectifier Company.

[17] “PWM Management for BLDC Motor Drives Using The ST72141,” AN1129 Application Note, STMicroelectronicsInc.

[18] 孫清華, 黃昌圳, 最新直流無刷馬達, 全華, 2001

[19] 鄭光耀, 無刷直流馬達無感測控制方法之研究與 DSP 實現技術之發展, 國立 交通大學電機與控制工程學系, 博士論文, 2003.

[20] 林穎燦, 無刷直流馬達無感測換相控制 IC 之規劃與設計, 國立交通大學電機 與控制工程系, 碩士論文, 2002.

[21] 林瀚宏, 無刷直流馬達之分析與無感測器驅動之建模, 國立交通大學電機與 控制工程系, 碩士論文, 2004.

[22] 吳欣達, 順滑模態估測器應用於無感測無刷直流馬達之速度控制, 國立交通 大學電機與控制工程系, 碩士論文, 2005.

Appendix A

Position detection and start-up algorithm of BLDC motors by using inductance variation

A method of identifying the rotor position of a BLDC motor and driving a motor smoothly from standstill without any position sensors is presented [18]. This is done by monitoring the current responses to the inductance variation on the rotor position.

Furthermore, the start-up algorithm also proposed to detect the next commutation timing by injecting a current pulse into six segments of an electrical cycle. In this section, this start-up method proposed by G. H. Jang, J. H. Park and J. H. Chang will be introduced in detail.

A.1 Position detection

The six segments of an electrical cycle are shown in Table A.1. In (3-28)-(3-32) represents that the phase current shows a different response depending on the inductance variation, which is determined by the relative position of a rotor and the direction of the current. Thus, the position information can be obtained by monitoring the phase current

i

⁺ and

i

⁻ in an appropriate time interval.

Table A.1 Six segments of an electrical cycle

Segment Symbol of current

B

In Fig.A.1, the current responses of the well-known six segments are calculated with the determined inductance in an appropriate interval of electrical degrees.

Observing Fig.A.1, it should be noted that the polarity of the difference between the positive current and negative current, , changes every 180 electrical degrees.

Furthermore, the inherent differences with 120 electrical degrees exist in any two phases such that the polarity of one of three changes every 60 electrical degrees, which can provide the position information, as shown in Fig.A.2. However, the position information from is not proper to drive a motor. The polarity of the difference between , , can be effectively used to identify the rotor position due to

it includes the information of the polarity of the difference between two back-EMF used in six-step drive. The polarity of one of three changes every 60 electrical degrees with 30 electrical degrees shift compared with the variation of , as shown in Fig. A.3 and Table A.2. Thus, the initial rotor position can be detected by monitoring

∆i

n

the polarity of , after each of six segments has been supplied a voltage pulse with a 20

∆∆i

k

µs period.

Fig. A.1 The current responses of the well-known six segments

Fig. A.2 The differences between the current

i

_n⁺ and

i

_n⁻

Fig. A.2 The differences between the current

i

_n⁺ and

i

_n⁻

Fig. A.3 The differences between the current

∆i

_n

Table A.2 Polarity of

∆∆i

_k on rotor position

Electrical position

∆∆ i

₁

∆∆i

₂

∆∆i

₃

30^ο ~ 90^ο + + −

90^ο ~ 150^ο + − −

150^ο ~ 210^ο + − +

210^ο ~ 270^ο − − +

270^ο ~ 330^ο − + +

330^ο ~ 30^ο − + −

A.2 Start-up algorithm

Once the stationary position of a rotor is identified, the correct phases of a BLDC motor are energized to produce the maximum torque. However, since six pulses cannot be inject during a commutation period as the motor moving quickly and some of these pulses generate negative torque by using position detection, the start-up algorithm is needed to detect the next commutation timing.

Consider the torques produced by the six segments with respect to the electrical angle as shown in Fig.A.4, in which every two phases out of three are energized in the positive or negative direction. Except the exciting segment which produces the maximum torque, there are still two segments to produce the torque with same direction in every commutation period. It should be noted that the increasing one is the next excitation. In other words, the increasing segment should replace the original exciting segment to be the new exciting segment while the torque produced by the increasing segment is larger than the torque produced by the original exciting segment.

For example, the phases A should be switched o the phases B A when a rotor C passes S_CB. The phases A is excited when the initial position is detected as 0~30 B electrical degrees. It means that the exciting phases is A and increasing phases is B

C

A . In Fig.A.5, P_phase and P_pulse correspond to the long and short pulses to energize the current the current phase A and the next commutation B A before the rotor passes C

S_CB. Therefore, the next excitation can be determined while the current of A is C larger than A the current of. The composition of the exciting phases and increasing B phases with respect to electrical angle is shown in Table A.3. The pulse train can accelerate the motor to medium speed. Once the motor produced sufficient back-EMF, the sensorless algorithm of the BLDC motor is switched to a back-EMF position detection method by using the zero-crossing of the back-EMF.

Fig.A.4 The torque generation of the six segments

Fig.A.5 The pulse train and its response

Table A.3 The exciting and increasing segment at electrical angle

Electrical position Exciting Increasing

30^ο ~ 90^ο A C BC

90^ο ~ 150^ο BC BA

150^ο ~ 210^ο BA CA

210^ο ~ 270^ο CA CB

270^ο ~ 330^ο C B AB

330^ο ~ 30^ο A B AC

在文檔中 直流無刷馬達無感測驅動技術之實現 (頁 74-0)