JP7626609B2 - Laser diode power supply unit - Google Patents

Description

This invention relates to a power supply device for driving a high-output, high-speed laser oscillator that is applied to metal processing processes such as welding and cutting. For example, it relates to a power supply device for driving a laser diode processing system that uses a laser diode as a laser oscillator and irradiates a target object with a laser beam (hereinafter, LD power supply).

The LD drive power supply drives the laser diode used as a pulsed laser oscillator at high speed. This requires a steep rise in the drive current, and the rise time is set, for example, in the range of 60 μs to 600 μs. In addition, the target value of the drive current and the pulse characteristics (DC output to 2 kHz) are also set to various values depending on the application.

The LD driving power supply is equipped with a power conversion unit including a switching element, and a control unit including PWM control with a feedback circuit, etc., so that the driving current has any characteristic. The control unit controls the pulse width of the PWM signal according to the target value of the driving current set by the user (Patent Document 1).

JP 2013-240801 A

However, in the above device, there is a problem that the rise time of the current waveform does not become optimal when the target value of the drive current fluctuates significantly due to the step response (transient response characteristic) of the rise of the drive current, which is determined by the transfer function of the control system including the control unit and the feedback unit. For example, if the rise step response is optimal under the condition that the target value of the drive current is 100% of the rated value, if the target value of the drive current drops significantly to about 25% of the rated value, the delay in the step response of the control system will become large. Here, the optimal rise step response means that the rise time until the drive current settles within a range of ±5% of the target value without overshooting is the fastest. If the delay in the step response of the control system becomes large, the rise time will be slower than necessary, and the time to reach the target value will also be longer. This is because when the target value drops significantly, the loop gain (hereinafter referred to as gain), which is the amplification factor of the transfer path of the control system, is insufficient compared to the rated 100%.

Conversely, if the gain is set so that the step response is optimal when the target value of the drive current is about 25% of the rated value, the gain will be excessive when the target value is 100% of the rated value, causing the rising waveform of the drive current to overshoot.

In this way, if the gain of the control system is set uniquely in a power supply device whose target value fluctuates greatly, it may not be possible to obtain the desired step response.

The object of this invention is to provide a power supply device for driving an LD that always optimizes the step response of the drive current by appropriately correcting the current command value of the laser diode drive current.

The power supply device for driving an LD of this invention includes a setting unit that sets information indicating the slope of the rise of the driving current and a target value of the driving current, a control unit that generates a current command value based on the information, and a power conversion unit that generates the driving current based on the current command value, and the control unit includes a correction unit that corrects the current command value from when it reaches a preset switching point until it reaches the target value.

The control unit generates a PWM signal based on the information until the current command value reaches a preset switching point. The drive current output from the power conversion unit is input to the laser diode.

The correction unit corrects the current command value when the current command value reaches a preset switching point. As a result, even if the target value fluctuates, the rising waveform of the drive current from the switching point to the target value is appropriate by setting the correction amount to a value corresponding to the target value. In other words, it is possible to prevent a slow rise time and the occurrence of overshoot.

The switching point is changed according to the target value. As the target value falls below the rated value, the deviation between the target value and the switching point becomes smaller. If this deviation becomes smaller than a certain level, it becomes difficult to control the current command value after the switching point. Therefore, it is desirable to also change the switching point when the target value falls below the rated value by more than a certain level.

The correction unit corrects the current command value based on the deviation between the target value and the current command value.

The calculation of the correction amount is simple and places a small burden on the control unit, since it only involves determining the deviation between the target value and the current command value. If the control unit is configured with a CPU, the current command value is obtained at each predetermined sampling. Therefore, the deviation amount is also calculated at each sampling.

The correction unit also corrects the current command value by multiplying the deviation amount by a predetermined correction amplification factor.

The correction amount is calculated not only by determining the deviation between the target value and the current command value, but also by multiplying the deviation by a predetermined correction amplification factor to adjust the peak value of the correction signal.

The correction unit can also suppress abrupt changes in the correction amount by passing the current command value through a first-order lag filter.

A first-order lag filter makes the step response curve gentler, preventing overshoot.

The amount of correction can be optimized using the correction amplification factor and first-order lag filter, making it possible to optimize the curve of the rising waveform until the drive current reaches the target value.

In addition, the information and the target value of the drive current can be sent from a higher-level system controller.

For example, a STEP signal that indirectly indicates the information is sent from a higher-level system controller. The setting unit has a table showing the correspondence between STEP signals and parameters indicating the slope, and obtains the parameter corresponding to the STEP number sent from the higher-level system from the table. This parameter becomes the slope information for controlling the current command value.

According to this invention, even if the target value fluctuates, the current command value can be appropriately corrected to optimize the rise time of the drive current from the switching point to the target value.

1 is a configuration diagram of a laser diode processing apparatus according to an embodiment of the present invention. 1 is a configuration diagram of a laser diode system including a power supply device 2 for driving an LD. Table T is shown. FIG. 2 is a block diagram of a control unit 21. 4 shows the step response characteristic of the current command value D. 4 is a flowchart showing the operation of a control unit 21. FIG. 13 is a diagram for clearly illustrating a method for determining a correction command value D'. This is an embodiment in which the correction amount R is adjusted. This is an embodiment in which the adjustment of the correction amount R is made more practical. Shows a first-order lag filter

Figure 1 is a diagram showing the configuration of a laser diode processing device according to an embodiment of the present invention.

The laser diode processing device is composed of a host system controller 1, a power supply device for driving a laser diode (hereinafter referred to as a LD driving power supply device) 2, a laser diode (hereinafter referred to as an LD) 3, and a processing head 4 that is fiber-connected to the LD 3. The LD 3 includes multiple laser diodes. The LD driving power supply device 2 and the LD 3 constitute a laser oscillator.

The upper system controller 1 outputs a STEP signal indirectly indicating the slope of the current command value D and a target value P of the drive current to the LD drive power supply device 2 before driving the LD 3. In the following explanation, when indicating the calculation period (sampling period), n is added. For example, the current command value D(n) indicates the current command value at the time of the current calculation, and the current command value D(n-1) indicates the current command value at the time of the previous calculation.

The control unit 21 in the LD driving power supply device 2 refers to table T to obtain a parameter C corresponding to the STEP signal. The control unit 21 controls the current command value D for each sampling by calculation using the parameter C described below.

The host system controller 1 also continuously outputs the target value P of the drive current to the LD drive power supply device 2 as an analog signal. The control unit 21 in the LD drive power supply device 2 acquires this target value P for each sampling. The host system controller 1 also outputs a start-up signal to the LD drive power supply device 2.

After receiving the start-up signal, the LD driving power supply unit 2 is driven by a PWM signal using a three-phase AC power supply as a power source, and outputs a driving current for the LD 3. The LD 3 is composed of a combination of multiple laser diode elements. The LD 3 irradiates a high-power laser beam 5 onto the workpiece 6 via the processing head 4. The LD driving power supply unit 2 has a large capacity of over 20 kW and generates a large current, so the laser beam 5 has the power to melt the beam spot on the workpiece 6.

Figure 2 is a diagram showing the configuration of a laser diode system including a power supply unit 2 for driving an LD.

The LD driving power supply device 2 includes a power conversion unit 20 that rectifies the three-phase AC voltage, controls the switching element SW with a PWM signal, and outputs the signal via a reactor L and a capacitor C. The LD driving power supply device 2 also includes a control unit 21 that generates a PWM signal, and a Hall CT 22 that detects the output current and feeds it back to the control unit 21.

The control unit 21 includes a setting unit 21a that sets information for controlling the rising slope of the current command value D (as described below, this information is parameter C) and a target value P of the current command value D, and a PWM generating unit 21b that generates a PWM signal and outputs it to the power conversion unit 20.

The upper system controller 1 outputs a start signal to start the LD driving power supply device 2 via the signal line 7. It also sends a STEP signal once to the control unit 21, and further continuously outputs the target value P. The STEP signal is information that indicates the slope of the rise of the driving current.

When the control unit 21 receives the STEP signal, it sets the parameter C by referring to table T in the setting unit 21a. Parameter C is a value for controlling the slope of the rise of the current command value D. As described below, in this embodiment, the STEP signal sent from the upper system controller 1 is set to 16 different rise times between 60 μs and 600 μs. The rise time of the current command value D is determined by the user in the upper system controller 1 by referring to the rated current and rated voltage of the LD3, etc.

Figure 3 shows table T.

Table T shows the correspondence between the STEP signal and the parameter C. The parameter C is a value that represents the slope of the current command value D (ΔI/Δt: current change rate for each sampling). In this example, there are 16 types of STEP signals, from 1 to 16, and the smaller the numerical value, the larger the value of the parameter C (the steeper the slope). For example, when the STEP signal is 1, the parameter C is 100, and parameter C=100 indicates that the slope of the current command value D is maximum, i.e., the rise time of the current command value D is shortest. Also, for example, when the STEP signal is 16, the parameter C is 50, and parameter C=50 indicates that the slope of the current command value D is minimum, i.e., the rise time of the current command value D is longest.

Note that the parameter C shown in Table T is an example value.

The control unit 21 sets the parameter C obtained by referring to table T in the setting unit 21a as a value for controlling the rising slope of the current command value D. The control unit 21 also sets the target value P of the drive current sent from the upper system controller 1 in the setting unit 21a. Unlike the STEP signal that is sent only once at the beginning, the target value P is sent continuously from the upper system controller 1.

When the control unit 21 receives a STEP signal, a target value P, and a start signal from the upper system controller 1, it extracts the parameter C stored (set) in the setting unit 21a and acquires the target value P for each sampling period. The control unit 21 controls the rising slope of the current command value D based on the acquired target value P. The PWM generation unit 21b and an error amplifier 21e (described later) generate a PWM signal based on the current command value D and output it to the power conversion unit 20.

The magnitude of the drive current output from the power conversion unit 20 is detected by the Hall CT 22 and fed back to the control unit 21. The PWM generation unit 21b controls the pulse width of the PWM signal so that the fed back detection value becomes the current command value D.

Figure 4 is a block diagram of the control unit 21.

The STEP signal sent from the upper system controller 1 is received by the setting unit 21a. A parameter C is extracted from the STEP signal by referring to table T, and the parameter C is set in the setting unit 21a. In addition, the target value P that is continuously sent from the upper system controller 1 is also set in the setting unit 21a.

The gradient control unit 21c reads out the parameter C set in the setting unit 21a as a digital value and controls the gradient of the rise of the current command value D. The gradient control unit 21c includes a correction unit 21d and corrects the current command value D.

The output of the gradient control unit 21c is output to the error amplifier 21e as a current command value D. The output of the Hall CT 22 is also input to the error amplifier 21e. The error amplifier 21e outputs the error between the current command value D and the Hall CT 22 to the PWM generation unit 21b. The PWM generation unit 21b and the error amplifier 21e control the pulse width of the PWM signal so that the error becomes zero.

The functions of the tilt control unit 21c and the correction unit 21d are realized by a CPU and software. The calculation period (CPU timer interrupt period) of the tilt control and correction control in the CPU is set to a high speed (for example, about 10 μs).

Next, we will explain the tilt control and correction control.

Figure 5 shows the step response characteristics of the current command value D.

In the figure, the solid line is the trajectory of the current command value D calculated for each sampling. The current command value D is calculated based on the parameter C set in the setting unit 21a. The time when the current command value D starts to rise is T1, the time of the switching point d is T2, and the time when the current command value D reaches the target value P and becomes stable is T3. The rise time is the time it takes for the current value indicated by the current command value D to reach 10% to 90% of the target value P. The switching point d is the position where the current value is 80% of the target value P when the load is not light. This switching point d can be arbitrarily determined by the designer in the control unit 21, and is automatically adjusted in the control unit 21 based on the magnitude of the fluctuation of the target value P (the deviation amount of the current). The switching point d is referred to as the timing at which the slope control is switched from the control between T1 and T2 (processing 1) to the control between T2 and T3 (processing 2).

The target value P is read from the setting unit 21a at the sampling period by the gradient control unit 21c. Also, in FIG. 5, the hatched area indicates the correction amount R for the current command value D. The correction amount R is the integral value of the correction value S for each sampling.

Process 1 is a process for controlling the slope of the current command value D between T1 and T2. In this process 1, the correction value S for each sampling from the correction unit 21d is zero.

Process 2 is a process for controlling the slope of the current command value D between T2 and T3. In process 2, the current command value D is controlled so as to reach the target value P gradually without overshooting. In this embodiment, in process 2, when the target value P is 100%, the step response characteristic of the current rise is optimal, so the correction amount R is set to zero. As the target value P decreases from there, a process is performed in which the correction amount R for the current command value D is increased. For example, if the target value P is 25% of the rated value, a correction amount R of 75% (100% - 25%) is added to the current command value D by the correction unit 21d.

The correction unit 21d is provided with a unique numerical table for obtaining the correction value S, and if the current rise time becomes gentler in response to the upper STEP signal, the correction value S is reduced or set to zero.

When the target value P falls below the rated value, in process 2, the corrected current command value (hereinafter, the corrected command value) D' = current command value D + correction value S is calculated for each calculation cycle. The correction value S is calculated as correction value S = (target value P - current command value D), and the hatched area in Figure 5 is the overall correction amount R for the current command value D between T2 and T3.

The characteristic curve between T2 and T3 from the switching point d to the target value P depends on the amplification factor (gain) of the transmission path of the analog control system including the error amplifier 21e, part of the PWM generating unit 21b, and the Hall CT. Therefore, this gain is usually calibrated so that the step response characteristic curve is optimal when the target value P is rated at 100%. The optimal current rise curve is a curve in which the current command value D reaches the target value P as quickly as possible without overshooting and stabilizes. On the other hand, the optimal value of the gain differs depending on the size of the target value P. In particular, when the target value P is significantly lower than the rated value (for example, 25% or less), the gain is insufficient. As a result, the current rise time until time T3 in FIG. 4 becomes long.

In this embodiment, as the target value P decreases from 100%, the correction unit 21d calculates the correction command value D' = current command value D + correction value S in process 2 as described above. This essentially increases the gain of the analog control system. The correction amount (area) R of the hatched region between T2 and T3 in Figure 5 is added to the current command value D shown by the solid line.

The above correction is selective. For example, if the target value P is 100% of the rated value, no correction is performed. If the target value P is equal to or lower than the rated value, the above correction is performed. This value is merely an example, and it is desirable to determine the target value P for correction by experimentation based on the characteristics of the transmission path of the analog control system and the characteristics of LD3.

As another example, the correction value S can be found by calculating the correction command value D' = current command value D + k (target value P - current command value D). k (target value P - current command value D) is the correction value S, and the coefficient k is varied depending on the magnitude of the target value P. For example, when the target value P is 100% of the rated value, k = 0, and when the target value P is 99% to 1% of the rated value, k is varied from 0.01 to 1.00 depending on the target value and added to the command value.

Figure 6 is a flowchart showing the operation of the control unit 21.

The operation period shown in this flowchart is 10 μs in this example, and this operation period is the calculation period (CPU timer interrupt period) during which the slope control unit 21c obtains the target value P, etc. from the setting unit 21a and controls the slope of the current command value D.

In ST1, the digital value of the target value P is obtained from the setting unit 21a. In order to avoid sudden changes and improve accuracy, this value is converted into a digital value by taking a moving average with the value obtained at the previous execution (sampling).

In ST2, the deviation DF between the target value P and the previous current command value D(n-1) is calculated.

The deviation amount DF is used in ST4.

In ST3, the setting unit 21a references table T to obtain parameter C, and also sets the switching point d (see Figure 5). As described above, switching point d is the position where the current value is 80% of the target value P when the load is not light. If the target value P becomes low and the current change amount is 2% or less of the rated ratio, it is fixed at 50% of the target value P. Specifically, it is appropriately set in advance by experiments, etc. The setting position of switching point d is changed because if the current change amount is low, the current difference between switching point d and target value P becomes small, and process 2 cannot be controlled well.

In ST4, the deviation amount DF is compared with the switching point d. The switching point d in ST4 is shown as a difference from 100% in order to compare with the deviation amount DF. That is, when the switching point d is at the 80% position, the switching point d in ST4 is expressed as 100%-80%=20%. If the deviation amount DF is less than the switching point d, the period is T1-T2, so proceed to process 1 (ST5), and if the deviation amount DF is equal to or greater than the switching point d, the period is T2-T3, so proceed to process 2 (ST6).

Process 1 is a process that calculates the current command value D(n) during the current rise time corresponding to the step signal between T1 and T2 in Figure 5. The current current command value D(n) is calculated using the following formula.

D(n)=f(C, D(n-1), DF)
Here, C is the parameter C acquired in ST3, and D(n-1) is the current command value output in the previous calculation. DF is the sum of the target value P and D(n-1) acquired in ST2. 1) is the deviation from

Process 2 is a process for calculating a pre-correction current command value D that reaches the target value as quickly as possible while suppressing overshoot between T2 and T3 in Figure 5. In process 1, the current command value D is controlled to have a linear slope, and in process 2, the pre-correction current command value D is controlled to reach the target value P with an appropriate curve. In process 2, the current pre-correction current command value D is calculated using the following formula.

D(n)=f'(D(n-1))
In the above formula, the function f represents a straight line, but the function f' is a logarithmic function that reaches the target value P with an appropriate curve.

Next, we will explain the correction process of correction unit 21d from ST7 onwards.

ST7 to ST10 are correction processes.

In ST7, the deviation amount DF is compared with the switching point d. As in ST4, in ST7, the switching point d is shown as a difference from 100% for comparison with the deviation amount DF. If the deviation amount DF is less than the switching point d, proceed to ST8, and if the deviation amount DF is equal to or greater than the switching point d, proceed to ST9.

In ST8, since it is the period T1-T2, the correction value S is zero.

In ST9, since it is the period T2-T3, the correction value S is calculated.

The correction value S is calculated from the current command value D(n) calculated in step 2 of ST6 and the target value P. In simple terms, it can be calculated as correction value S(n) = target value P - current command value D(n).

After completing the correction processes in ST7-ST9, in ST10, the corrected command value D(n)' is calculated by D(n)' = D(n) + S. The corrected command value D(n)' is the current command value obtained by this correction.

Note that no correction is made in ST8 during the T1-T2 period, so correction is only made in process 2.

ST11 is a process for updating the current command value D(n) obtained in the calculation of ST5 or ST6 and the corrected command value D(n)' obtained in the calculation of ST10 to the current current command value D(n) and the current corrected command value D(n)'. The current command value D(n) updated in ST11 is used as D(n-1) in the next calculation process.

By the above control, the current command value D is updated at the calculation cycle. In addition, in process 2, the current command value D is corrected based on the magnitude of the target value P (a corrected command value D' is obtained).

Figure 7 clearly shows how to calculate the correction value S.

The original value of the correction amount R, which is the integral value of the correction value S, can be found by repeating the calculation in ST10. The correction amount R is inverted to find an inverted value R', and this inverted value R' is added to the current command value D after the switching point d.

In this way, since the calculation of the correction value S in ST10 is simple, the correction command value D' in ST11 can also be easily calculated. Therefore, the burden on the tilt control unit 21c is light.

Figure 8 shows an example of adjusting the correction amount R.

In FIG. 8, the correction amplification factor of the transmission path of the analog control system is manipulated when calculating the correction amount R. That is, the correction value S is calculated by multiplying the deviation amount by the correction amplification factor. When the correction amplification factor is increased, the area of the correction amount R (the integral value of the correction value S) increases. When the correction amplification factor is decreased, the area of the correction amount R (the integral value of the correction value S) decreases. The correction amplification factor is set by the amplifier of the correction unit 21d, and its value is determined by experiment.

Figure 9 shows an example of a more practical adjustment of the correction amount R.

9, the correction value S is filtered with a first-order lag filter when calculating the correction amount R. By filtering the correction value S with a first-order lag filter, the rise of the correction amount R becomes gentler and easier to operate, resulting in a more practical correction value S.

The first-order lag filter is simply constructed using resistors and capacitors. For example, it can be constructed using an RC series circuit as shown in Figure 10. The amount of filtering can be set arbitrarily by adjusting the size of the resistor and capacitor, and the value can be obtained through experimentation. The first-order lag filter is actually realized by software in the gradient control unit 21c, which is made up of a CPU.

As described above, by adjusting the correction amplification rate and inserting a first-order lag filter, the correction amount R can be made more appropriate when the target value P fluctuates more than the rated value.

1--Higher system controller 2--Power supply device for driving laser diode 3--Laser oscillator 21--Control unit 21c--Inclination control unit 21d--Correction unit

Claims (3)

1. A power supply device for driving a laser diode, the power supply device setting a target value of a pulsed driving current every time the laser diode is outputted with the pulsed driving current,
a control unit that receives and sets information for controlling a rise gradient of a current command value from an external device, receives and sets a target value of the pulsed drive current for each of the pulsed drive currents from an external device, generates a current command value based on the information, and generates a PWM signal so that an error between the current command value and the pulsed drive current output to the laser diode becomes zero;
a power conversion unit that generates the pulsed drive current based on an output of the control unit,
The control unit is
changing a switching point as the target value, which is set every time the pulsed driving current is output, falls below a rated value;
a first process is performed in a first period in which the switching point is reached, the first process is for generating the current command value based on information for controlling the slope, and a second process is performed in a second period in which the switching point is reached and the target value is reached, the second process is for correcting the current command value based on a deviation amount between the current command value based on information for controlling the slope and the target value;
increasing a correction amount of the current command value in the second process as the target value decreases below the rated value.
Power supply for driving laser diodes. The power supply device for driving a laser diode according to claim 1, wherein the correction is performed by multiplying the deviation amount of the current command value by a predetermined correction amplification factor. 3. The power supply device for driving a laser diode according to claim 1, wherein said information and said target value of said pulsed driving current are sent from a host system controller.

Images