KR101135253B1 - Method for sensing motor speed - Google Patents

Abstract

PURPOSE: A method for detecting the speed of a motor is provided to measure the speed of the motor without an error even if a new pulse is inputted while measuring the speed of the motor. CONSTITUTION: A first M value corresponding to transition counts of an phase signal and a B phase signal is saved in a first reference transition time which is reference waveform transition timing(S1). A first T value offered from a timer is provided to the first reference transition time(S2). The timer counts corresponding to the transition counts of the phase signal and the B phase signal. A second M value is saved in a second reference transition time(S3). A second T value offered from the timer is provided to the second reference transition time(S4). The speed of a motor is obtained as the first M value and a first T value in case the first M value and the second M value are same. The speed of the motor is obtained as the second M value and the second T value in case the first M value and the second M value are different.

Description

Motor speed detection method {Method for sensing Motor Speed}

The present invention relates to a speed measuring method using an encoder during motor operation. In particular, the present invention relates to a method of measuring speed by using a signal provided by an encoder in an M / T method.

In general, the method of measuring the speed of a motor with an encoder output pulse during motor operation using an encoder is as follows. First, there is a method of counting the number of pearls coming in for a certain time. This is called the M method. There is also a method of measuring the time between pulses. This is called the T method. The method of using the M method and the T method together is called the M / T method. Among them, the best performance method is to measure motor speed required for motor operation using M / T method.

FIG. 1 is a waveform diagram showing an M method, a T method, and an M / T method, respectively.

The M method uses the following formula (RPM = (60 * m * 2π) / (2π * Tc * Ppr), Ppr: encoder pulses per revolution, Tc: measurement cycle, m: pulses during Tc) You will get it. The pulse counted for a predetermined time may be obtained through Equation 1 below. The POSCNT address of FIG. 1A is a storage address that stores a counted number of times for a predetermined time set in a TMS320F2833X series DSP chip.

Figure 1b is a diagram when the T method is implemented with a DSP of the TMS320F2833X series. In FIG. 1B, the QCTMR address is a storage address that counts and stores a time between pulses. The stored time can be obtained by the following formula (Rpm = 60 / (△ T * Ppr), Ppr: number of encoder pulses per revolution, ΔT: time between pulse and pulse).

Figure 1c is a general diagram for the M / T method. The M / T method is a method of measuring by using the number of encoder pulses input for a predetermined time and the input time of the pulses. That is, using the pulse m for a predetermined time and ΔT for this pulse m, the speed can be obtained through the following equation (3) (RPM = 60 * m / (Ppr * ΔT)).

The DSP chip of the TMS320F2833X series described above is equipped with an encoder module that is easy to use by a user. However, the matters that can be changed by the user are greatly reduced, and thus, only the M and T methods can be used as shown in FIGS. 1A and 2B.

However, in the case of the M method, there is a shortcoming in the low speed due to lack of information in the low speed, in the case of the T method, there is a disadvantage in that it is difficult to accurately measure the time between the pulse at high speed and frequent operation at high speed. .

The present invention is to provide a method that can easily perform the M / T method to compensate for the disadvantages of the M and T method. In addition, it provides a M / T method that can measure the speed without error even if a new pulse is input during the speed measurement.

The present invention provides a method of generating a reference waveform having a transition timing corresponding to reference transition time points of an A-phase signal and a B-phase signal provided with a 90-degree phase from a camcorder to measure a speed of a motor; Storing a first M value corresponding to the number of transitions of the A-phase signal and the B-phase signal at a first reference transition time, which is a timing at which the reference waveform transitions; Providing a first T value provided by a timer that is counted corresponding to the number of transitions of the A-phase signal and the B-phase signal at the reference transition time point; Storing a second M value corresponding to the number of transitions of the A-phase signal and the B-phase signal at a second reference transition time point that is after a predetermined timing from the first reference transition time point; Providing a second T value provided by a timer counting corresponding to the number of transitions of the A-phase signal and the B-phase signal at the second reference transition time point; Comparing the first M value with the second M value and obtaining the speed of the motor using the first M value and the first T value; And comparing the first M value with the second M value to obtain the speed of the motor using the second M value and the second T value.

The predetermined timing may be smaller than a half period of the reference waveform.

The reference waveform having a transition timing corresponding to a reference transition time point of the A-phase signal and the B-phase signal may be generated by performing an exclusive OR logic on the A-phase signal and the B-phase signal.

According to the present invention, the speed of an encoder can be easily measured in an M / T method. In addition, the speed can be measured without error even if a new pulse is input during the speed measurement.

1 is a waveform diagram showing various methods of obtaining speed using an encoder;
2 is a rotating disc and an operating waveform diagram of an optical encoder.
3 is a waveform diagram showing the operation of an encoder;
4 is a waveform diagram showing a state of an encoder signal;
FIG. 5 is a diagram showing a change in state and a change in a counter according to the waveform shown in FIG. 4; FIG.
Fig. 6 is a diagram showing an encoder processing truth table.
7 is an internal block diagram of a control unit for measuring the speed of an encoder according to an embodiment of the present invention.
FIG. 8 is a waveform diagram showing the operation of the control unit shown in FIG. 7; FIG.
FIG. 9 is a block diagram showing the connection and operation of an encoder and a controller shown in FIG.
Fig. 10 is a block diagram showing a capture register.
Fig. 11 is a waveform diagram showing a method of measuring the speed of an encoder according to the present invention.
Fig. 12 is a block diagram showing hardware required for implementing the M / T method of the present invention and the method.
Fig. 13 is a flowchart showing a method of measuring the speed of an encoder according to the present invention.

DETAILED DESCRIPTION Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the technical idea of the present invention. do.

The present invention proposes a method for implementing an M / T method using a DSP provided by TI of the TMS320F2833X series. Compared to the M and T methods, the M / T method can calculate the motor speed through encoder signals more accurately and efficiently.

However, since the TMS320F2833X does not have a latch function, the M / T method is not provided by default. The TMS320F2833X has a capture function, which is one of the other functions of the TMS320F2833X, to implement the M / T method. This M / T implementation using the TMS320F2833X series of DSP chips can reduce costs and reduce PCB area by eliminating additional hardware such as FPGAs. Although the gain from implementing M / T using only the TMS320F2833X family of chips without additional hardware is one of the cores of the patent, the present invention can also be implemented in a DSP chip having a capture function rather than the DSP chip of the TMS320F2833X.

The encoder is attached to the shaft of the rotary motor. When the motor rotates, the encoder shaft rotates together. When the power is supplied to the encoder, it sends out signals called A and B phases.

2 is a rotating disc and an operating waveform diagram of an optical encoder. 3 is a waveform diagram showing an operation of an encoder. 4 is a waveform diagram showing a state of an encoder signal. FIG. 5 is a diagram illustrating a change of state and a change of a counter according to the waveform shown in FIG. 4. Fig. 6 is a diagram showing an encoder processing truth table.

As shown in Fig. 2, the encoder has a slotted slot on a rotatable disk, and the back of the disk projects light through an LED or the like. Rotating the disk can produce signals for QEPA, QEPB, and QEPI depending on when light is passing through or not. In order to generate the signals of QEPA and QEPB, two discs are used so that the slots overlap only half. Then, the order of the QEPA and QEPB signals is changed according to the rotation direction. In addition, the period of the pulses of the QEPA and QEPB changes according to the rotation speed. Through this, both the rotation direction and the rotation speed of the motor can be detected. The encoder specification is called PPR (Pulse Per Revolution). This value indicates how many pulses are emitted per revolution. Larger values allow more accurate position and velocity calculations.

The Enhanced Quadrature Encoder Pulse (eQEP) module refers to a hardware function that allows the speed of the motor to be calculated using the output of the encoder's A and B phase signals. Continuing, we explain the principle of counting pulses through the eQEP module. 3 shows how the encoder signal, the pulse counter, and the direction signal change. When QDIR = 1, the motor rotates in the forward direction, and when QDIR = 0, the motor rotates in the reverse direction. QPOSCNT shows the change in pulse counter. +1 indicates that the pulse counter is increased by 1 and -1 is the pulse counter is decreased by 1.

We will now explain how to determine the increase or decrease of QPOSCNT. In FIG. 4, when the encoder output signals QEPA and QEPB have the same value of 0 and 1, the state is assigned to 00, 01, 10, 11. Fig. 4 shows the allocation according to the state of the waveform, and Fig. 5 shows the change of state and the increase and decrease of the counter. 6 shows an encoder processing truth table.

7 is an internal block diagram of a controller for measuring the speed of an encoder according to an embodiment of the present invention. FIG. 8 is a waveform diagram illustrating the operation of the controller illustrated in FIG. 7.

Referring to Figures 7 and 8, the eCAP module provided in the TMS320F2833X series DPS chip receives the encoder signal through this processing and stores it in a register called QPOSCNT (a type of memory device in which data is stored in the DSP chip). Reading the value will tell you how much the motor has rotated.

This article explains why M / T cannot be easily implemented in the TMS320F2833X series. In general, in order to easily implement M / T, a capture function should be provided. The capture function saves the current value of a register to a temporary memory when a certain event occurs (the moment a condition is met).

In the M / T method, a specific event is the moment when the pulse counter changes, and the values to be moved to the temporary storage place are the M value and the T value. The M value uses a pulse counter register like QPOSCNT and the T value uses the timer value at the moment the pulse changes. A free running timer means a timer that continuously increases with a constant clock. When the timer reaches its maximum value, it is reset to zero. In the case of a 16-bit timer, it increments up to 65535 and then goes back to 0 and starts incrementing.

With capture registers, the POSCNTCAP and TCAP values can be used to accurately calculate the average speed from the time of the previous capture to the moment of the current capture. Two capture registers must be operated in one set. An event should occur with only the POSCNTCAP value read and the TCAP should not change to the new value. To prevent this malfunction, the capture enable / disable function is required. While disabled, the capture register's value remains unchanged even if an event occurs.

In the TMS320F2833X series of DSP chips, the free-runner timer is automatically reset when a capture event occurs, as shown in Figure 8. Multiple events can occur within the speed measurement interval (speed measurement interrupt, usually 1ms) .If the free runner timer is reset each time an event occurs, the time from the start event to the last event is increased from the timer capture value read in the last event. You will not know. Therefore, it is difficult to implement M / T method in the existing way. Due to this difficulty, the present embodiment uses the eCAP function of the TMS320F2833X to implement timer capture in another way.

FIG. 9 is a block diagram illustrating a connection between a controller and an encoder illustrated in FIG. 7 and an operation thereof. 10 is a block diagram showing a capture register. 11 is a waveform diagram showing a method of measuring the speed of an encoder according to the present invention. 12 is a flowchart showing a method of measuring the speed of an encoder according to the present invention.

Next, the eCAP function of the TMS320F2833X series will be described with reference to FIGS. 9 to 12. The eCAP function captures the value of a specific register when a specific event occurs and stores it in another register (called a capture register). Here, the event can be set to be the edge of the signal input to the pin from the TMS320F2833X series of DSP chips. To use this function, the circuit is constructed as shown in FIG. Connect A, B phase signal to A, B input of eQEP module. Connect A-phase and B-phase signals to the inputs of the XOR logic gates and connect their outputs to the inputs of the eCAP module. By doing so, the counter value and the capture value included in the TMS320F2833X series DSP chips may not change at the input of the eCAP module due to the signal processing delay of the XOR gates of the A pulse and the B pulse. The treatment for such a case is also one of the cores of the patent. The maximum value at which the pulse can enter is assumed by the maximum value of the motor speed. If a 1024 PPR encoder is attached to a 100,000 rpm motor, 10000/60 Hz (1024 ⅹ 4) pulse / sec = 6826667 pulse / sec, and the inverse would be 0.146 us / pulse.

After reading the counter QPOSCNT that stores the location information, if the new counter pulse is received while reading the CAP register that stores the time information, the CAP register changes. In this case, the time information is wrong and the M / T calculation is wrong. To avoid this, if the counter pulse is read again and this value is different from the previously read value, the CAP register may be wrong. Therefore, the newly read CAP register value is used instead of this value.

In order to measure the speed of motor by M / T method, eQEP module and eCAP module are used because of chip structure of TMS320F2833X series. However, the eCAP on the TMS320F2833X family of chips does not have a latching function, requiring the encoder count to be read once more. When an interrupt corresponding to the speed control cycle (= measurement cycle) occurs, the encoder count is latched and stored in the Position_Latch (QPOSLAT register). However, since ΔT is not latched, the eCAP capture values are read in sequence immediately after the interrupt occurs. As a new pulse comes in, the capture value is incremented so the count reads Real_position (QPOSCNT register), not the value latched into Position_Latch (QPOSLAT register).

When the measurement cycle interrupt occurs, immediately after the value of position_1 (QPOSCNT register), CAP_1 (four registers from CAP1 to 4), position_2 value (QPOSCNT register), and CAP_2 (four registers from CAP1 to 4) In order to read these two values, the CAP_1 value is taken. If the location values are different, the CAP_2 value is taken. The reason for obtaining the same function value twice is to eliminate the error that occurs when a new pulse comes in while reading the register for speed measurement as described above.

In other words, when the measurement period is reached as shown in Figure 4, when the pulse is input as shown in (3) in Figure 4, it is recognized that one pulse is increased and is higher than the actual speed (the number of pulses is increased but If the information is not updated), the measurement is made low (time information is updated but the pulse is not updated).

Therefore, in order to measure no error even if the pulse like (3) in Figure 4 is input, measure twice and if the number of pulses is changed, take the 2_CAP value as described above and remove the error.

The measured position value is calculated as the difference (△ M) from the existing position counter.

The method for obtaining the pulse time DELTA T is as follows.

Acquire the largest value among CAP1 ~ 4 among the selected CAP values. The CAP value thus obtained becomes a new time, and the pulse time is obtained by a difference from the existing time. Using ΔM and ΔT thus obtained, the motor speed can be obtained as shown in Equation 4 (RPM = 60 * m / (Ppr * ΔT)).

Through the present invention, it is possible to detect the motor speed of the M / T method even in the DSP base of the TMS320F2833X series. Accordingly, even in the DSP-based TMS320F2833X series, motor speed detection can be measured very accurately in all areas from low speed to high speed, and speed calculation can be performed without any error even if a new pulse is input during measurement. In addition to measuring motor speed, it can be used in devices that generate commands using pulses such as MPG (Manual pulse generator). It is also possible to use it as a speed command of the motor to be controlled by using pulses generated by other devices, such as motor synchronous operation.

Figure 12 is a block diagram showing the hardware required for implementing the M / T method of the present invention and the method.

In order to obtain the M / T value according to the present embodiment, a logic gate 62 for exclusively ORing the A-phase signal and the B-phase signal provided to the encoder 60, the M value providing unit 61, and the calculation unit 64. ), A count 63.

The count is counted from the moment the A-phase signal and the B-phase signal are output up to a predetermined value. The M value providing unit 61 receives the A-phase signal and the B-phase signal, and stores and outputs the M value according to the number of transitions thereof. The logic gate 62 outputs a reference waveform that transitions in response to the reference transition time of the A-phase signal and the B-phase signal. 12, waveforms of the A-phase signal, the B-phase signal, and the output signal of the logic gate 62 are shown.

The calculation unit 64 includes two registers, and the first M value provided by the M value providing unit 61, the second M value, and the count are provided at the time points k1 and k2 at which the speed is measured. The first counting value and the second counting value are stored.

If the two M values input to the calculation unit 64 are equal to each other and are the same value, the speed of the M / T method is obtained using the first counting value and the first M value input first. If it is not the same value, the speed of the M / T method is obtained using the second counting value and the second M value.

The reason why the M and T values are inputted and compared with each other twice (k1, k2) at the time of measuring the speed is that when the reference waveform changes at the time of measuring the speed, new M and T values are provided. By using the newly provided value, the speed of the M / T method can be obtained, so that accurate speed can be measured.

Fig. 13 is a flowchart showing a method of measuring the speed of an encoder according to the present invention.

Referring to FIG. 13, a speed measuring method according to the present embodiment has a transition timing corresponding to reference transition time points of the A-phase signal and the B-phase signal provided with a 90-degree phase from a camcorder to measure the motor speed. Generate a reference waveform.

Subsequently, a first M value corresponding to the number of transitions of the A-phase signal and the B-phase signal is stored at a first reference transition time, which is a timing at which the reference waveform transitions (see S1).

Subsequently, a first T value provided by a timer counted corresponding to the number of transitions of the A-phase signal and the B-phase signal is provided at the reference transition point (see S2). Here the counting value is the T value.

Subsequently, a second M value corresponding to the number of transitions of the A-phase signal and the B-phase signal is stored at a second reference transition time point, which is after a predetermined timing from the first reference transition time point (see S3).

Subsequently, a second T value provided by a timer counted corresponding to the number of transitions of the A-phase signal and the B-phase signal at the second reference transition point is provided (see S4).

Subsequently, when the first M value and the second M value are equal to each other (see S5), the first M value and the first T value are selected to determine the speed of the motor (S6, S8, and S9). .

Subsequently, if the first M value and the second M value are not equal to each other (see S5), the speed of the motor is obtained using the second M value and the second T value (S7, S8, and S9). .

In addition, the predetermined timing is characterized in that less than one-half period of the reference waveform.

The reference waveform having a transition timing corresponding to a reference transition time point of the A-phase signal and the B-phase signal may be generated by performing an exclusive OR logic on the A-phase signal and the B-phase signal.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is clearly understood that the same is by way of illustration and example only and is not to be construed as limiting the scope of the present invention. I will understand. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined by the scope of the appended claims, as well as the appended claims.

Claims (3)

Generating a reference waveform having a transition timing corresponding to reference transition time points of the A-phase signal and the B-phase signal provided with a 90-degree phase from the camcorder to measure the speed of the motor;
Storing a first M value corresponding to the number of transitions of the A-phase signal and the B-phase signal at a first reference transition time, which is a timing at which the reference waveform transitions;
Providing a first T value provided by a timer that is counted corresponding to the number of transitions of the A-phase signal and the B-phase signal at the reference transition time point;
Storing a second M value corresponding to the number of transitions of the A-phase signal and the B-phase signal at a second reference transition time point that is after a predetermined timing from the first reference transition time point; And
Providing a second T value provided by a timer counting corresponding to the number of transitions of the A-phase signal and the B-phase signal at the second reference transition time point;
Comparing the first M value with the second M value and obtaining the speed of the motor using the first M value and the first T value; And
And comparing the first M value with the second M value to obtain the speed of the motor using the second M value and the second T value.
The method of claim 1,
And said predetermined timing is smaller than one-half period of said reference waveform.
The method of claim 1,
A reference waveform having a transition timing corresponding to the reference transition time point of the A-phase signal and the B-phase signal is
And generating the A-phase signal and the B-phase signal by performing exclusive OR logic.

Images