Why is the encoder phase of permanent magnet AC servo motor aligned with the rotor magnetic pole phase
Its sole purpose is to achieve the goal of vector control, decoupling the d-axis excitation component and q-axis output component, so that the electromagnetic field generated by the stator winding of the permanent magnet AC servo motor is always orthogonal to the rotor permanent magnetic field, thereby obtaining the best output effect, that is, the "DC like characteristic". This control method is also known as field oriented control (FOC). The external manifestation of achieving the FOC control goal is that the "phase current" waveform of the permanent magnet AC servo motor is always consistent with the "opposite potential" waveform, as shown in the following figure:

Therefore, it can be inferred that as long as a way is found to keep the "phase current" waveform of the permanent magnet AC servo motor consistent with the "opposite potential" waveform, the FOC control goal can be achieved, so that the primary electromagnetic field of the permanent magnet AC servo motor is orthogonal to the magnetic pole permanent magnetic field, that is, there is a 90 degree electrical angle difference between the waveforms, as shown in the following figure:

How to find a way to keep the "phase current" waveform of permanent magnet AC servo motor consistent with the "opposite potential" waveform? As shown in Figure 1, as long as the electrical angle phase of the sinusoidal back electromotive force waveform can be detected at any time, it is relatively easy to generate a sinusoidal phase current waveform that is consistent with the back electromotive force waveform based on the electrical angle phase.
It should be noted that the so-called electrical angle of a permanent magnet AC servo motor is the sine (Sin) phase of the opposite potential waveform of phase a (U), so phase alignment can be converted into the alignment relationship between the encoder phase and the back electromotive force waveform phase; On the other hand, the electrical angle is also the angle between the d-axis (straight axis) of the rotor coordinate system and the a-axis (U-axis) or alpha axis of the stator coordinate system, which is helpful for graphical analysis.
In practical operation, European and American manufacturers are accustomed to aligning the phase of the encoder and rotor magnetic poles by directing the motor rotor with a DC current smaller than the rated current through the winding of the motor. When a DC current less than the rated current is applied to the winding of the motor, under the condition of no external force, the primary electromagnetic field and the magnetic pole permanent magnetic field will interact and attract each other, and be positioned at the equilibrium position with a phase difference of 0 degrees, as shown in the following figure:

Comparing Figure 3 and Figure 2 above, it can be seen that although the position of the a-phase (U-phase) winding (red) is at the peak center (specific angle) of the electromagnetic field waveform, under FOC control, the center of the a-phase (U-phase) is aligned with the q-axis of the permanent magnet; When oriented without load, the center of phase A (phase U) is aligned with the d-axis. That is to say, compared to the primary (stator) winding, the d-axis of the secondary (rotor) magnet coordinate system will shift 90 degrees to the left during no-load orientation, coinciding with the original position of the q-axis under FOC control. This achieves the alignment relationship between the a-axis (U-axis) or α - axis and the d-axis during rotor no-load orientation.
At this point, the phase is aligned to an electrical angle of 0 degrees, and the direction of the rotor directional current applied in the motor winding is BC phase (VW phase) in and A phase (U phase) out. Due to the parallel relationship between B phase (V phase) and C phase (W phase), the current flowing through B phase (V phase) and C phase (W phase) may be unbalanced, thereby affecting the accuracy of rotor orientation.
The practical method for applying directional current to rotors is to input phase b (V) and output phase a (U), that is, to connect phase a (U) and phase b (V) in series, which can obtain phase a (U) and phase b (V) currents with completely consistent amplitudes, which is beneficial for the accuracy of orientation. At this time, the position of the phase a (U) winding (red) is 30 degrees away from the d-axis, that is, the axis a (U) or α is aligned to the electrical angle position that is 30 degrees away from the d-axis (negative), as shown in the figure:

The waveform of the opposite potential and the line back potential corresponding to the two rotor orientation methods mentioned above, as well as the relationship between the electrical angle, are shown in the following figure. The brown line represents the alignment of the a-axis (U-axis) or α - axis with the d-axis, which is directly aligned to the electrical angle 0 point; The purple line is aligned with the a-axis (U-axis) or a-axis to an electrical angle position that is 30 degrees (negative) away from the d-axis, that is, aligned to the -30 degree electrical angle point:

The solid brown line in the figure indicates that the d-axis is aligned with the a-axis (U-axis) or alpha axis, that is, aligned to the electrical angle 0 point. The alignment method is to apply a current vector with a fixed phase angle of -90 degrees to the motor winding, as shown by the brown dashed line in the figure. Under no-load conditions, the d-axis of the motor rotor will move towards the position of the q-axis component of the current vector with a phase angle of -90 degrees under FOC control, which coincides with the a-axis or a-axis in the figure, and is ultimately oriented at that position, that is, the electrical angle of 0 degrees.
The d-axis shown by the purple solid line is 30 degrees away from the a-axis (U-axis) or alpha axis, aligning to the -30 degree electrical angle point. The alignment method is to apply a current vector with a fixed phase angle of -120 degrees to the motor winding. Under no-load conditions, the d-axis of the motor rotor will move towards the position of the q-axis component of the current vector with a phase angle of -120 degrees in FOC, which is 30 degrees clockwise from the a-axis or a-axis in the figure, and finally oriented at that position, that is, the electrical angle of -30 degrees.
Explanation: The descriptions of U, V, W phases and a, b, c phases in the text have a one-to-one correspondence with the U, V, W axes and a, b, c axes.
The mainstream servo motor position feedback components include incremental encoders, absolute encoders, sine and cosine encoders, rotary transformers, etc.
Phase alignment method of incremental encoder
In this discussion, the output signal of incremental encoders is a square wave signal, which can be divided into incremental encoders with commutation signals and ordinary incremental encoders. Ordinary incremental encoders have two-phase orthogonal square wave pulse output signals A and B, as well as a zero position signal Z. Incremental encoders with commutation signals not only have ABZ output signals, but also electronic commutation signals UVW that are 120 degrees different from each other. The number of cycles per revolution of UVW is consistent with the number of magnetic pole pairs of the motor rotor. The alignment method between the phase of the UVW electronic commutation signal of the incremental encoder with commutation signal and the rotor magnetic pole phase, or electrical angle phase, is as follows:
1. Use a DC power supply to supply the UV winding of the motor with a DC current lower than the rated current, V in and U out, and orient the motor shaft to a balanced position;
2. Observe the U-phase signal and Z-signal of the encoder with an oscilloscope;
3. Adjust the relative position between the encoder shaft and the motor shaft;
4. While adjusting, observe the transition edge of the encoder U-phase signal and the Z signal until the Z signal stabilizes at a high level (assuming the normal state of the Z signal is low), and lock the relative position relationship between the encoder and the motor;
5. Twist the motor shaft back and forth. After letting go, if the Z signal stabilizes at a high level every time the motor shaft returns freely to the equilibrium position, the alignment is effective.
After removing the DC power supply, verify the following:
1. Observe the U-phase signal of the encoder and the UV back electromotive force waveform of the motor using an oscilloscope;
2. Rotate the motor shaft counterclockwise, and the rising edge of the U-phase signal of the encoder coincides with the zero crossing point of the UV line back electromotive force waveform of the motor from low to high. The Z signal of the encoder also appears at this zero crossing point.
The above verification method can also be used as an alignment method.
It should be noted that at this point, the phase zero of the incremental encoder's U-phase signal is aligned with the phase zero of the motor's UV line back electromotive force. Due to the opposite potential of the motor's U, there is a 30 degree difference between the phase zero of the incremental encoder's U-phase signal and the UV line back electromotive force. Therefore, after this alignment, the phase zero of the incremental encoder's U-phase signal is aligned with the -30 degree phase point of the motor's U-opposite potential, and the motor's electrical angle phase is consistent with the phase of the U-opposite potential waveform. Therefore, at this point, the phase zero of the incremental encoder's U-phase signal is aligned with the -30 degree point of the motor's electrical angle phase.
Some servo companies are accustomed to aligning the zero point of the encoder's U-phase signal directly with the zero point of the motor's electrical angle. To achieve this goal, they can:
1. Use a DC power supply to apply a DC current lower than the rated current to the UVW winding of the motor, with VW input and U output, and orient the motor shaft to a balanced position;
2. Observe the U-phase signal and Z-signal of the encoder with an oscilloscope;
3. Adjust the relative position between the encoder shaft and the motor shaft;
4. While adjusting, observe the transition edge of the encoder U-phase signal and the Z signal until the Z signal stabilizes at a high level (assuming the normal state of the Z signal is low), and lock the relative position relationship between the encoder and the motor;
5. Twist the motor shaft back and forth. After letting go, if the Z signal stabilizes at a high level every time the motor shaft returns freely to the equilibrium position, the alignment is effective.
The verification method is as follows:
1. Connect three resistors with equal resistance values into a star shape, and then connect the three resistors connected in the star shape to the UVW three-phase winding leads of the motor;
2. By observing the midpoint between the motor's U-phase input and the star shaped resistor with an oscilloscope, the waveform of the motor's U-opposite potential can be approximately obtained;
3. Rotate the motor shaft counterclockwise, and it can be seen that the rising edge of the encoder's U-phase signal coincides with the zero crossing point of the motor's U opposite potential waveform from low to high.
The above verification method can also be used as an alignment method.
Due to the lack of UVW phase information in ordinary incremental encoders, and the fact that the Z signal can only reflect one point within a circle, it does not have the potential for direct phase alignment and is therefore not the topic of this discussion.
Phase alignment method of absolute encoder
The phase alignment of absolute encoders is not significantly different for single and multiple cycles, as they align the detected phase of the encoder with the phase of the motor's electrical angle within one cycle. Early absolute encoders used a separate pin to provide the highest bit level of a single loop phase. By flipping the 0 and 1 of this level, phase alignment between the encoder and the motor can also be achieved. The method is as follows:
1. Use a DC power supply to supply the UV winding of the motor with a DC current lower than the rated current, V in and U out, and orient the motor shaft to a balanced position;
2. Observe the highest digital level signal of the absolute encoder with an oscilloscope;
3. Adjust the relative position between the encoder shaft and the motor shaft;
4. Adjust while observing the transition edge of the highest count digit signal until the transition edge accurately appears at the directional balance position of the motor shaft, and lock the relative position relationship between the encoder and the motor;
5. Twist the motor shaft back and forth. After letting go, if the jumping edge can be accurately reproduced every time the motor shaft returns to the equilibrium position freely, then the alignment is effective.
This type of absolute encoder has been widely replaced by serial protocols such as EnDAT, BiSS, Hyperface, as well as new absolute encoders using Japanese specific serial protocols. Therefore, the highest bit signal no longer exists, and the method of aligning the encoder and motor phase has also changed. One very practical method is to use the EEPROM inside the encoder to store the measured phase of the encoder randomly installed on the motor shaft. The specific method is as follows:
1. Randomly install the encoder on the motor, that is, fix the encoder shaft and motor shaft, as well as the encoder housing and motor housing;
2. Use a DC power supply to supply the UV winding of the motor with a DC current lower than the rated current, V in and U out, and orient the motor shaft to a balanced position;
3. Use a servo driver to read the single turn position value of the absolute encoder and store it in the EEPROM that records the initial phase of the motor's electrical angle inside the encoder;
4. The alignment process is completed.
Since the motor shaft is now oriented in the -30 degree direction of the electrical angle phase, the position detection value stored in the internal EEPROM of the encoder corresponds to the -30 degree phase of the motor electrical angle. Afterwards, the driver will subtract the single turn position detection data at any time from this stored value, and perform necessary conversions based on the number of motor poles, plus -30 degrees, to obtain the motor electrical angle phase at that time.
This alignment method requires the support and cooperation of encoders and servo drivers to achieve. The fundamental reason why the encoder phase of Japanese servos is not convenient for end users to adjust directly is that they refuse to provide users with the functional interface and operation method of this alignment method. One major advantage of this alignment method is that it only requires providing the motor winding with a rotor oriented current that determines the phase sequence and direction, without adjusting the angular relationship between the encoder and the motor shaft. Therefore, the encoder can be directly installed on the motor at any initial angle without the need for fine or even simple adjustment processes, with simple operation and good processability.
If an absolute encoder has neither EEPROM available for use nor the highest count digit pin available for detection, the alignment method will be relatively complex. If the drive supports the reading and display of single coil absolute position information, then it can be considered:
1. Use a DC power supply to supply the UV winding of the motor with a DC current lower than the rated current, V in and U out, and orient the motor shaft to a balanced position;
2. Use a servo drive to read and display the single turn position value of the absolute encoder;
3. Adjust the relative position between the encoder shaft and the motor shaft;
4. After the above adjustments, make the displayed absolute position value of a single turn fully close to the absolute position point of the motor corresponding to the -30 degree electrical angle calculated based on the number of pole pairs of the motor, and lock the relative position relationship between the encoder and the motor;
5. Twist the motor shaft back and forth. After letting go, if the motor shaft can accurately reproduce the above converted position points every time it returns to the equilibrium position freely, then the alignment is effective.
If the user cannot even obtain absolute value information, they can only rely on the original factory's dedicated tooling to detect the absolute position detection value while detecting the motor electrical angle phase. Using the tooling, adjust the relative angular position relationship between the encoder and the motor, align the encoder phase with the motor electrical angle phase, and then lock them. In this way, users have even more difficulty solving the phase alignment problem of the encoder on their own.
My personal recommendation is to use the method of storing the initial installation position in EEPROM, which is simple, practical, adaptable, and easy to open to users for them to install the encoder and complete the phase setting of the motor's electrical angle.
Phase alignment method of sine cosine encoder
A regular sine cosine encoder has a pair of orthogonal sin, cos 1Vp-p signals, which are equivalent to the AB orthogonal signals of an incremental encoder for square wave signals. Each cycle repeats many signal cycles, such as 2048; And a narrow amplitude symmetrical triangular wave Index signal, equivalent to the Z signal of an incremental encoder, usually appearing once per circle; This sine cosine encoder is essentially an incremental encoder. Another type of sine cosine encoder not only has the orthogonal sin and cos signals mentioned above, but also has a pair of mutually orthogonal 1Vp-p sine type C and D signals that appear only one signal cycle per circle. If the C signal is used as sin, then the D signal is cos. Rotate the encoder shaft counterclockwise, which is equivalent to aligning the index signal of the Z signal with the zero crossing point of the C signal from low to high. Through the high magnification subdivision technology of sin and cos signals, not only can the sine and cosine encoder obtain a finer nominal detection resolution than the original signal period, for example, a 2048 line sine and cosine encoder can achieve a nominal detection resolution of over 4 million lines per revolution after 2048 subdivision. Currently, many European and American servo manufacturers provide such high-resolution servo systems, while domestic manufacturers are not yet common; In addition, the C and D signals of the sine and cosine encoder with C and D signals can provide high absolute position information per revolution after subdivision, such as 2048 absolute positions per revolution. Therefore, the sine and cosine encoder with C and D signals can be regarded as an analog single turn absolute encoder.
The initial electrical angle phase alignment method of the servo motor using this encoder is as follows:
1. Use a DC power supply to supply the UV winding of the motor with a DC current lower than the rated current, V in and U out, and orient the motor shaft to a balanced position;
2. Observe the waveforms of the C signal and Index signal of the sine cosine encoder using an oscilloscope;
3. Adjust the relative position between the encoder shaft and the motor shaft;
4. Adjust while observing the waveforms of the C signal and Index signal until the zero crossing point of the C signal or the effective level of the Index signal accurately appears at the directional balance position of the motor shaft, locking the relative position relationship between the encoder and the motor;
5. Twist the motor shaft back and forth. After letting go, if the zero crossing point of the C signal or the effective level of the Index signal can be accurately reproduced each time the motor shaft returns to the equilibrium position, then the alignment is effective.
After removing the DC power supply, verify the following:
1. Observe the C-phase signal of the encoder and the UV back electromotive force waveform of the motor using an oscilloscope;
2. Rotate the motor shaft counterclockwise, and the zero crossing point of the encoder's C-phase signal from low to high or the jump edge of the Index signal coincides with the zero crossing point of the motor's UV line back electromotive force waveform from low to high.
This verification method can also be used as an alignment method.
At this point, the zero crossing point of the C signal is aligned with the -30 degree point of the motor electrical angle phase.
If you want to align directly with the 0-degree point of the motor's electrical angle, you can consider:
1. Use a DC power supply to apply a DC current lower than the rated current to the UVW winding of the motor, with VW input and U output, and orient the motor shaft to a balanced position;
2. Observe the C signal and Index signal waveforms of the encoder using an oscilloscope;
3. Adjust the relative position between the encoder shaft and the motor shaft;
4. Adjust while observing the waveforms of the C signal and Index signal until the zero crossing point of the C signal or the effective level of the Index signal accurately appears at the directional balance position of the motor shaft, locking the relative position relationship between the encoder and the motor;
5. Twist the motor shaft back and forth. After letting go, if the zero crossing point of the C signal or the effective level of the Index signal can remain stable at a high level every time the motor shaft returns to the equilibrium position freely, then the alignment is effective.
The verification method is as follows:
1. Connect three resistors with equal resistance values into a star shape, and then connect the three resistors connected in the star shape to the UVW three-phase winding leads of the motor;
2. By observing the midpoint between the motor's U-phase input and the star shaped resistor with an oscilloscope, the waveform of the motor's U-opposite potential can be approximately obtained;
3. Rotate the encoder shaft counterclockwise and observe that the zero crossing point of the encoder's C-phase signal from low to high or the transition edge of the Index signal should coincide with the zero crossing point of the motor U's opposite potential waveform from low to high.
The above verification method can also be used as an alignment method.
Due to the fact that ordinary sine cosine encoders do not have phase information within one circle, and the Index signal can only reflect one point within one circle, it does not have the potential for direct phase alignment. Therefore, it is not a topic of discussion here.
If the servo drive that can be connected to the sine cosine encoder can provide users with single turn absolute position information obtained from C and D, then we can consider:
1. Use a DC power supply to supply the UV winding of the motor with a DC current lower than the rated current, V in and U out, and orient the motor shaft to a balanced position;
2. Use a servo drive to read and display the absolute position information of a single turn obtained from the C and D signals;
3. Adjust the relative position between the rotary shaft and the motor shaft;
4. After the above adjustments, make the displayed absolute position value sufficiently close to the absolute position point corresponding to the -30 degree electrical angle of the motor calculated based on the number of pole pairs of the motor, and lock the relative position relationship between the encoder and the motor;
5. Twist the motor shaft back and forth. After letting go, if the motor shaft can accurately reproduce the converted absolute position points each time it returns to the equilibrium position, then the alignment is effective.
Afterwards, after removing the DC power supply, the same alignment verification effect as before can be obtained:
1. Observe the C-phase signal of the sine cosine encoder and the UV back electromotive force waveform of the motor using an oscilloscope;
2. Rotate the motor shaft to verify that the zero crossing point of the encoder's C-phase signal from low to high coincides with the zero crossing point of the motor's UV line back electromotive force waveform from low to high.
If non-volatile memory such as EEPROM inside the driver is used, the measured phase of the sine cosine encoder randomly installed on the motor shaft can also be stored. The specific method is as follows:
1. Randomly install sine and cosine on the motor, that is, fix the encoder shaft and motor shaft, as well as the encoder housing and motor housing;
2. Use a DC power supply to supply the UV winding of the motor with a DC current lower than the rated current, V in and U out, and orient the motor shaft to a balanced position;
3. Use a servo drive to read the absolute position value of a single turn analyzed from the C and D signals, and store it in non-volatile memory such as EEPROM that records the initial installation phase of the motor's electrical angle inside the drive;
4. The alignment process is completed.
Since the motor shaft is now oriented in the -30 degree direction of the electrical angle phase, the position detection values stored in non-volatile memory such as EEPROM inside the driver correspond to the -30 degree phase of the motor's electrical angle. Afterwards, the driver will subtract the absolute position value of a single turn related to the electrical angle analyzed by the encoder at any time from this stored value, and perform necessary conversions based on the number of motor poles. Adding -30 degrees, the motor electrical angle phase at that time can be obtained.
This alignment method requires support and cooperation from the domestic and operational aspects of the servo drive in order to achieve it. Moreover, since non-volatile memory such as EEPROM that records the initial phase of the motor's electrical angle is located in the servo drive, once aligned, the motor is actually bound to the drive. If the motor, sine cosine encoder, or drive needs to be replaced, the initial installation phase alignment operation needs to be performed again, and the matching relationship between the motor and the drive needs to be re bound.
Phase alignment method of rotary transformer
Rotary transformer, abbreviated as rotary transformer, is composed of high-performance silicon steel laminations and enameled wires that have undergone special electromagnetic design. Compared with encoders using optoelectronic technology, it has heat resistance and vibration resistance. The adaptability to harsh working environments such as impact resistance, oil resistance, and even corrosion resistance makes it widely used in weapon systems and other harsh working conditions. A pair of pole (single speed) rotary transformers can be regarded as a single coil absolute feedback system, and are also the most widely used. Therefore, only single speed rotary transformers are discussed here. Multi speed rotary transformers are matched with servo motors, and I personally believe that the number of pole pairs is best to use the divisor of the motor pole pairs, which facilitates the correspondence of motor degrees and the decomposition of pole pairs.
The signal lead of the rotary transformer is generally 6 wires, divided into 3 groups, corresponding to one excitation coil and 2 orthogonal induction coils. The excitation coil receives the input sine excitation signal, and the induction coil induces a detection signal with SIN and COS envelope based on the mutual angular position relationship of the rotary transformer stator. The output signals of the rotating SIN and COS are modulated based on the angle between the rotor and stator. If the excitation signal is sin ω t and the electrical angle between the rotor and stator is θ, then the SIN signal is sin ω t × sin θ, and the COS signal is sin ω t × cos θ. Based on the SIN, COS signals and the original excitation signal, high-resolution position detection results can be obtained through necessary detection circuits. Currently, the detection resolution of commercial rotating systems can reach the 12th power of 2 per turn, which is 4096, while scientific research and aerospace systems can even reach the 20th power of 2 or more, but the volume and cost are also very considerable.
Here, it is assumed that as the rotary rotor CCW rotates, the electrical angle phase of the rotary transformer increases, and as the rotary rotor CW rotates, the electrical angle phase of the rotary transformer decreases.
The alignment method for the electrical angle phase of commercial rotary transformers and servo motors is as follows:
1. Use a DC power supply to supply the UV winding of the motor with a DC current lower than the rated current, V in and U out;
2. Then use an oscilloscope to observe the signal lead output of the rotating SIN coil;
3. Adjust the relative position between the rotary rotor on the motor shaft and the motor shaft, or the relative position between the rotary stator and the motor housing, based on the convenience of operation;
4. Adjust while observing the envelope of the rotating SIN signal until the amplitude of the signal envelope completely returns to zero, and lock the rotating signal;
(4‘). Adjust while observing the Lissajous diagram with the Sin signal as the horizontal axis and the excitation signal as the vertical axis, until the Lissajous diagram becomes a perpendicular line that coincides with the vertical axis, and twist the perpendicular line towards quadrants 1 and 3 in the CCW direction and towards quadrants 2 and 4 in the CW direction to lock the rotation;
5. Twist the motor shaft back and forth. After letting go, if the amplitude zero crossing of the signal envelope can be accurately reproduced every time the motor shaft returns to the equilibrium position freely, or if Lissajous can coincide with the vertical axis as a perpendicular line, then the alignment is effective.
Remove the DC power supply and perform alignment verification:
1. Observe the SIN signal of the rotary converter and the UV back electromotive force waveform of the motor with an oscilloscope;
2. Rotate the motor shaft to verify that the zero crossing point of the SIN signal envelope of the rotary transformer coincides with the zero crossing point of the UV line back electromotive force waveform of the motor from low to high.
This verification method can also be used as an alignment method.
At this point, the zero crossing point of the SIN signal envelope is aligned with the -30 degree point of the motor electrical angle phase.
If you want to align directly with the 0-degree point of the motor's electrical angle, you can consider:
1. Use a DC power supply to apply a DC current lower than the rated current to the UVW winding of the motor, with VW input and U output, and orient the motor shaft to a balanced position;
2. Observe the rotating SIN signal with an oscilloscope;
3. Adjust the relative position between the rotating shaft and the motor shaft;
4. Adjust while observing the envelope waveform of the SIN signal until the amplitude of the signal envelope completely returns to zero, and lock the rotation;
(4'). Adjust while observing the Lissajous diagram with the Sin signal as the horizontal axis and the excitation signal as the vertical axis, until the Lissajous diagram becomes a perpendicular line that coincides with the vertical axis, and twist the perpendicular line towards quadrants 1 and 3 in the CCW direction and towards quadrants 2 and 4 in the CW direction to lock the rotation;
5. Twist the motor shaft back and forth. After letting go, if the amplitude zero crossing of the signal envelope can be accurately reproduced every time the motor shaft returns to the equilibrium position freely, or if Lissajous can coincide with the vertical axis as a perpendicular line, then the alignment is effective.
The verification method is as follows:
1. Connect three resistors with equal resistance values into a star shape, and then connect the three resistors connected in the star shape to the UVW three-phase winding leads of the motor;
2. By observing the midpoint between the motor's U-phase input and the star shaped resistor with an oscilloscope, the waveform of the motor's U-opposite potential can be approximately obtained;
3. Use an oscilloscope to observe the zero crossing points of the SIN signal envelope and the zero crossing points of the opposite potential waveform of motor U from low to high. These two zero crossing points should coincide.
The above verification method can also be used as an alignment method.
It should be pointed out that in the above operation, it is necessary to effectively distinguish the positive and negative half cycles in the rotating SIN envelope signal. Due to the fact that the SIN signal is the modulation result of the excitation signal based on the sin θ value of the angle θ between the stator and rotor, in the SIN signal envelope corresponding to the positive half cycle of sin θ, the modulated excitation signal is in phase with the original excitation signal, while in the SIN signal envelope corresponding to the negative half cycle of sin θ, the modulated excitation signal is in phase opposition to the original excitation signal. Based on this, it is possible to distinguish and judge the positive and negative half cycles in the waveform of the SIN envelope signal output by the inverter. When aligning, it is necessary to take the zero crossing point of the SIN envelope signal corresponding to the transition point from the negative half cycle to the positive half cycle of sin θ. If it is taken in reverse or not accurately judged, the aligned electrical angle may be misaligned. 180 degrees, which may cause the speed outer loop to enter positive feedback.
If the servo drive that can be connected to the rotary transformer can provide users with absolute position information related to the motor electrical angle obtained from the rotary transformer signal, then it can be considered:
1. Use a DC power supply to supply the UV winding of the motor with a DC current lower than the rated current, V in and U out, and orient the motor shaft to a balanced position;
2. Use a servo drive to read and display the absolute position information related to the motor electrical angle obtained from the rotary signal;
3. Adjust the relative position between the rotary transformer shaft and the motor shaft, or the relative position between the rotary transformer housing and the motor housing, based on the convenience of operation;
4. After the above adjustments, make the displayed absolute position value sufficiently close to the absolute position point corresponding to the -30 degree electrical angle of the motor calculated based on the number of pole pairs of the motor, and lock the relative position relationship between the rotor and the motor shaft;
5. Twist the motor shaft back and forth. After letting go, if the motor shaft can accurately reproduce the converted absolute position points each time it returns to the equilibrium position, then the alignment is effective.
Afterwards, after removing the DC power supply, the same alignment verification effect as before can be obtained:
1. Observe the SIN signal of the rotary converter and the UV back electromotive force waveform of the motor with an oscilloscope;
2. Rotate the motor shaft to verify that the zero crossing point of the SIN signal envelope of the rotary transformer coincides with the zero crossing point of the UV line back electromotive force waveform of the motor from low to high.
If non-volatile memory such as EEPROM inside the driver is used, the measured phase of the rotary transformer randomly installed on the motor shaft can also be stored. The specific method is as follows:
1. Randomly install the rotary transformer on the motor, that is, fix the rotary transformer shaft and motor shaft, as well as the rotary transformer housing and motor housing;
2. Use a DC power supply to supply the UV winding of the motor with a DC current lower than the rated current, V in and U out, and orient the motor shaft to a balanced position;
3. Use a servo drive to read the absolute position value related to the electrical angle analyzed by the rotary converter, and store it in non-volatile memory such as EEPROM that records the initial installation phase of the motor's electrical angle inside the drive;
4. The alignment process is completed.
Since the motor shaft is now oriented in the -30 degree direction of the electrical angle phase, the position detection values stored in non-volatile memory such as EEPROM inside the driver correspond to the -30 degree phase of the motor's electrical angle. Afterwards, the driver will subtract the absolute position value related to the electrical angle analyzed by the rotary transformer at any time from this stored value, and perform necessary conversions based on the number of motor poles. Adding -30 degrees, the motor electrical angle phase at that time can be obtained.
This alignment method requires support and cooperation from the domestic and operational aspects of the servo drive in order to achieve it. Moreover, non-volatile memory such as EEPROM that records the initial phase of the motor's electrical angle is located in the servo drive. Therefore, once aligned, the motor is actually bound to the drive. If the motor, transformer, or drive needs to be replaced, the initial installation phase alignment operation needs to be performed again, and the matching relationship between the motor and the drive needs to be re bound.
attention
In the above discussion, the term 'aligned to the motor electrical angle at a -30 degree phase' is based on the premise that the UV back electromotive force waveform lags behind the U phase by 30 degrees.
2. In the above discussion, VU is used for communication and the UV line back electromotive force waveform is taken as an example. Some servo systems may use UW communication and refer to the UW line back electromotive force waveform for alignment.
3. If you want to align directly to the 0 degree phase point of the motor electrical angle, you can also connect the U-phase to the negative end of the low-voltage DC source, and connect the V-phase and W-phase in parallel to the positive end of the DC source. At this time, the orientation angle of the motor shaft will be offset by 30 degrees relative to the series connection of the UV phase. After aligning with the corresponding alignment method given in the article, it will be aligned to the 0 degree phase of the motor electrical angle in principle, without any offset of -30 degrees. This may seem beneficial, but considering the inconsistency of motor winding parameters, the currents flowing through the V-phase and W-phase windings in parallel may not be consistent, which can affect the accuracy of the motor shaft orientation angle. When VU is electrified, the U-phase and V-phase windings are simply connected in series, so the current flowing through the U-phase and V-phase windings must be consistent, and the accuracy of the motor shaft orientation angle will not be affected by the winding orientation current.
4. It cannot be ruled out that servo manufacturers may intentionally align the initial phase misalignment, especially in feedback systems that can provide absolute position data. The misalignment alignment of the initial phase can be easily compensated for by the offset of the data, which may serve to protect their product line. However, in this way, users have no way of knowing where the initial phase of the servo motor feedback component should be aligned. Users naturally do not want to encounter such suppliers.
Summary of Basic Methods for Electrical Angle Phase Alignment
1. Waveform observation method
Suitable for incremental encoders, sine cosine encoders, and rotary transformers with commutation signals.
1) By directly observing the phase alignment relationship between the zero crossing point of the UV line back electromotive force waveform and the rising edge/Z signal of the U-phase signal, or the zero crossing point of the Sin signal, or the zero crossing point of the Sin envelope signal of the sensor using an oscilloscope, the above signal edges or zero crossings of the sensor can be aligned to a -30 degree electrical angle phase;
2) Three equivalent resistors with appropriate resistance range are used to form a star shape, which is connected to the UVW power line of the permanent magnet servo motor. The virtual U opposite potential waveform between the U-phase power line and the center point of the star equivalent resistor is observed with an oscilloscope, and the phase alignment relationship with the rising edge/Z signal of the U-phase signal, or the zero crossing point of the Sin signal, or the zero crossing point of the Sin envelope signal of the sensor is observed. This method can align the above signal edge or zero crossing point of the sensor to the electrical angle phase 0 point;
2. Rotor orientation method
Suitable for waveform alignment of incremental encoders, sine and cosine encoders, and rotary transformers with commutation signals, or absolute encoders and sine and cosine encoders, rotary transformers, etc. that can provide single turn absolute position numerical information alignment.
1) Connect the V-phase to the positive terminal of the low-voltage DC source and the U-phase to the negative terminal of the DC source, and orient the motor shaft
Afterwards, while adjusting the relative position relationship between the sensor and the motor, observe the sensor signal with an oscilloscope until the rising edge of the U-phase signal or the Z-signal, or the zero crossing point of the Sin signal, or the zero crossing point of the Sin envelope signal is accurately reproduced. This method can align the edge or zero crossing point of the sensor's signal to the -30 degree electrical angle phase;
It is also possible to adjust the relative position relationship between the sensor and the motor while trying to observe the numerical information of the absolute position of a single turn until the zero position of the data is accurately reproduced. This method can also align the zero point of the absolute position of a single turn of the sensor to the -30 degree electrical angle phase;
If the value of the absolute position of a single turn corresponding to a -30 degree electrical angle is estimated in advance, the relative position relationship between the sensor and the motor can be adjusted until the value is accurately reproduced. Then, the zero point of the absolute position of a single turn can be directly aligned to the zero point of the electrical angle phase (this method may have better accuracy than the latter method summarized in the next section 2);
Of course, it is also possible to simply install an encoder randomly without adjusting the relative position relationship between the sensor and the motor, and use the read absolute position information of a single turn as the initial bias value for installation. Through subsequent calculations, the logical alignment between the absolute position information of a single turn and the zero point of the electrical angle phase can be achieved. This method requires minimal manual operation.
2) Connect the U-phase to the negative pole of the low-voltage DC source, connect the V-phase and W-phase in parallel to the positive pole of the DC source, and orient the motor shaft
Afterwards, while adjusting the relative position relationship between the sensor and the motor, observe the sensor signal with an oscilloscope until the rising edge of the U-phase signal or the Z-signal, or the zero crossing point of the Sin signal, or the zero crossing point of the Sin envelope signal is accurately reproduced. This method can align the edge or zero crossing point of the sensor's signal to the electrical angle phase zero point;
It is also possible to adjust the relative position relationship between the sensor and the motor while trying to observe the numerical information of the absolute position of a single turn until the data zero position is accurately reproduced. This method can also align the above signal edges or zero crossings of the sensor to the electrical angle phase zero point.