JPH0923641A - Drive system for switching power supply - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize stabilized control while facilitating the regulation of load current to be born in the case of parallel operation of DC-DC converter system switching power supplies. SOLUTION: A charging circuit 1 charges a capacitor C1 and a comparator Q2 compares the charging voltage with a threshold level Vref1 responding to the load current with regard to an external switching pulse PLS1 for PWM control. Output from the comparator Q2 and the switching pulse PLS1 are inputted to the OR gate Q3 to produce a PWM control signal for switching power supply. More specifically, a predetermined pulse width (on-time) is added by the charging circuit 1 to the width of inputted switching pulse PLS1. Subsequently, a pulse width (on-time) responding to the load current is subtracted therefrom to produce a PWM control signal for switching control.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a switching power supply drive system suitable for parallel operation of a DC-DC converter system switching power supply.

[0002]

2. Description of the Related Art FIG. 7 is a diagram showing a circuit configuration of a general switching power supply. In the figure, E is a DC power supply, SW is a switch, L is a choke coil, D is a diode, and C is a smoothing capacitor.

Although the circuit operation described above is well known, details thereof will be omitted, but the output is controlled by PWM (pulse width modulation) by a switching pulse from the outside. At this time, as shown in FIG. 8, assuming that the ON / OFF cycle of the switch SW to be switching-controlled is T and the ON time is τon, the output voltage Eo across the smoothing capacitor C is Eo. = E
・ (Τon / T).

FIG. 9 shows the relationship between ON / OFF of the switch SW and the output voltage (smoothing capacitor voltage) Eo. The output voltage Eo is the pulse width (time width) of the ON time τon of the switch SW by PWM control. Changes.

When a plurality of blocks of the above switching power supplies are connected in parallel for parallel operation, the basic condition of the parallel connection is that the output characteristics of the blocks are the same. If the output characteristics do not match, the load currents of the blocks connected in parallel cannot share the same value, and a block that is unilaterally overloaded occurs, which is extremely dangerous for the entire device. It will be operated in a normal state.

[0006] For example, the conditions of parallel connection adopted in generators and the like are that the output voltage when there is no load is the same as the load characteristics (regulation curve). These conditions are the load sharing of each block. Is a condition for being uniform.

When the output voltage increases, the internal impedance of each block tends to decrease,
This is a natural result for reducing the circuit loss, and it is not a problem in itself. However, considering parallel connection, this decrease in internal impedance means that it becomes very difficult to adjust the load balance.

As described above, a necessary condition for operating a plurality of power supplies (blocks) in parallel is that the respective no-load voltages and internal impedances must have values close to each other. The reason for this is that even if the difference in the no-load voltage or the internal impedance is small as described above, the load currents of the respective blocks will be significantly different values, which is shown in FIG. .

For example, the no-load voltage of one block is set to V
1 and draw the load line of this block (the straight line showing the state of the decrease of the output voltage with respect to the output current). Draw the load line.

Now, assuming that these five blocks are connected in parallel, their output voltages Eo share the same value. Further, since the load currents of the blocks having different load lines are the current values when the voltage value of each load line is Eo, if the no-load voltage is different, the load lines may have moved parallel along the voltage axis. become.

Therefore, as shown in FIG. 10, each current is determined as I1, I2, I3, I4, I5. If such characteristics are actually present, the sharing of the load current is extremely unbalanced.

A technique for forming the core of a low loss, high output converter is a switching system.
The control method adopted to realize excellent controllability is the PWM control described above. At present, it is virtually impossible to realize a simple and highly practical control method other than this PWM.

The basic problem of the above switching method in the parallel connection of a plurality of blocks is that the width of the PWM pulse (switching pulse) is accumulated due to the delay of rising and falling during transfer. This delay time varies not only with the amount unique to each block, but also with changes in ambient temperature and is sensitive to changes in ambient conditions.

Further, in the usual case, a diode is used to prevent a circulating current from flowing between the smoothing capacitors of each block, but energy is accumulated in the inductance associated with the wiring of the diode, etc. An uncalculated voltage rise occurs, and this boosting effect makes it more difficult to adjust the load balance.

The most characteristic aspect of the boosting effect is that even if the blocks are manufactured in exactly the same way, if the load is unbalanced due to some cause caused by the circumstances on the load side,
The imbalance of loads may be further expanded. In reality, the condition under which this effect occurs is only the relationship between the amount of increase in voltage and the amount of decrease due to line loss. Normally, line loss often exceeds the amount of voltage increase, so it does not appear as a phenomenon.

Needless to say, some measures must be taken when the line loss cannot exceed a certain value due to heat generation or the like.

[0017]

By the way, in the conventional switching power supply drive system, when a plurality of blocks are operated in parallel as described above, it is difficult and stable to adjust the load current due to the imbalance of the load. There was a problem that it could not be controlled.

That is, as described above, the factors causing the load imbalance are the individual differences (no-load voltage and internal impedance) of each block, the delay time difference (PWM pulse width difference), and the boosting effect ( There is a difference in line impedance and line loss. Even if the characteristics of each block are made uniform by adjustment due to the phenomenon of boosting effect, the load balance may be destroyed when the load current exceeds a certain value.

The present invention has been made in view of the above problems, and provides a drive system of a switching power supply in which a load current can be easily adjusted during parallel operation and stable control can be performed. Is intended.

[0020]

A drive system of a switching power supply according to the present invention is a drive system of a switching power supply whose output is controlled by pulse width modulation, and a predetermined pulse width of a switching pulse given from the outside. The adding means for adding the ON time width of the above and the subtracting means for subtracting the ON time width according to the load current from the added pulse width are provided, and the switching control is performed by the subtracted pulse.

[0021]

According to the present invention, a constant pulse width is added to the pulse width given from the outside, an amount corresponding to the load current is subtracted from the added pulse width, and the subtracted pulse width is added. PWM control is performed by.

[0022]

FIG. 1 is a circuit diagram showing an embodiment of the present invention. In the figure, 1 is a charging circuit for charging the capacitor C1 with a constant current in response to a switching pulse PLS1 given from the outside, 2 is a discharging circuit for the capacitor C1,
Reference numeral 3 is a sensor for detecting a load current, and the detection signal thereof is input to the discharge circuit 2.

Q1 is a switching pulse PLS1
An inverter for inverting and inputting to the discharge circuit 2, a comparator Q2 for comparing the voltage of the capacitor C1 with a predetermined threshold voltage Vref1, and an inverter Q3 for inputting the output of the comparator Q2 and the switching pulse PLS1. At the (OR) gate,
The output controls the PWM of the switching power supply. ZD is a Zener diode connected in parallel with the capacitor C1.

Here, the charging circuit 1 constitutes an adding means for adding a predetermined ON time (pulse width) to the pulse width of the switching pulse PLS1, and the discharging circuit 2 adds a load based on the added pulse width. A subtraction unit is configured to subtract the on-time (pulse width) according to the current. Then, the switching control of the switching power supply is performed by the subtracted pulse.

Next, the operation will be described. Conventionally, there is no active means for balancing the load, and when the PWM operation is mainly used, the delay time of the pulse may exceed the tolerance, which makes it difficult to adjust the load current.

However, in the present embodiment, means for adjusting the load balance is provided, so that the load current can be easily adjusted when the switching power supplies are operated in parallel, and stable control is possible.

That is, a certain pulse width is added to the pulse width of the switching pulse PLS1 given from the outside, and the amount corresponding to the load current is subtracted from the added pulse width. As a result, the load current of the block falls within a certain variation.

The pulse width (tag pulse width) to be added is given in proportion to the pulse width of the switching pulse PLS1 from the outside when the pulse width is less than a certain value (minimum tag pulse).

The advantage of this method using the tag pulse is that when multiple switching power supplies are connected in parallel, each block automatically determines the optimum pulse width according to the load current that it shares, so that the load is extremely low. It is difficult for the imbalance of

FIG. 2 is a diagram showing the control operation of the above-described embodiment. In the circuit of FIG. 1, when the charging circuit 1 charges the capacitor C1 with a constant current I1 in response to a switching pulse PLS1 given from the outside, the capacitor C1
The voltage V1 of rises at a constant slope. At this time, the discharge circuit 2 is in a rest state.

At this time, if the voltage of the capacitor C1 is clipped so as not to exceed a certain value, a pulse as shown in FIG. 2A is obtained. Then, at the same time when one cycle of the switching pulse PLS1 ends, the electric charge accumulated in the capacitor C1 is discharged by the discharge circuit 2. At this time, the charging circuit 1 is in a dormant state.

For the above-mentioned discharge method and time setting, the discharge current is a sum of a certain constant value and a value proportional to the load current. The state is shown in FIG. The degree of decrease in the capacitor voltage during discharging is faster as the load current is larger and slower as the load current is smaller.

Further, the voltage across the capacitor C1 is monitored by the comparator Q2, and the pulse width obtained at that time is a pulse width according to the load current. That is, when the voltage across the capacitor C1 reaches a constant value (Vref1), the output of the comparator Q2 changes from low (Low) to high (High).
h), the pulse width is large if the load current is small, so that a large load current will be shared in the next cycle (cycle). On the contrary, if the load current is large, the pulse width is small, so that a small load current will be shared in the next cycle.

As described above, if the average value measuring operation at the time of detecting the load current is appropriate, the sharing of the load current is improved by several steps as compared with the prior art. 3 (a) to 3 (d) show output waveforms of each part of FIG. 1, and finally the gate output of the OR gate Q3 is sent to the switching control part of the block to drive the switching element such as FET. To be done.

FIG. 4 is a circuit configuration diagram showing another embodiment of the present invention, and the same reference numerals as those in FIG. 1 denote the same components.

In the embodiment of FIG. 1 described above, regarding the creation of the pulse width (generation of the tag pulse), "the method of discharging in proportion to the load current", that is, the discharge current of the capacitor C1 is a constant value + the value proportional to the load current. However, in the present embodiment, the discharge current is fixed and the threshold value for voltage comparison is made proportional to the load current.
Is adopted.

That is, the sensor 3 in the circuit of FIG. 4 detects the load current and inputs a voltage proportional to the load current to the comparator Q2 as the threshold voltage Vref2. Thereby, as shown in (a) to (c) of FIG. 5, the comparison level in the comparator Q2 is changed according to the magnitude of the load current, and when the load current is large, the pulse width becomes small and the load current When it is small, the pulse width is large. 6 (a) to 6 (d) show output waveforms of the respective parts of the circuit of FIG.

Even with such a configuration, the same operational effects as those of the above-described embodiment can be obtained, the load current can be easily adjusted during parallel operation, and stable control can be performed. If the control system as a whole is required to be linear, this embodiment is preferable, and the method of the above-mentioned embodiment may be combined.

[0039]

As described above, according to the present invention, the predetermined width is added to the pulse width of the PWM control given from the outside, and the amount corresponding to the load current is subtracted, so that the parallel operation is performed. When carrying out the above, there is an effect that the sharing of the load current can be easily adjusted and stable control can be performed.

[Brief description of drawings]

FIG. 1 is a circuit configuration diagram showing an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing the operation of one embodiment.

3 is an output waveform diagram of each part of the circuit of FIG.

FIG. 4 is a circuit configuration diagram showing another embodiment of the present invention.

FIG. 5 is an explanatory diagram showing the operation of another embodiment.

6 is an output waveform diagram of each part of the circuit of FIG.

FIG. 7 is a circuit configuration diagram showing an example of a general switching power supply.

FIG. 8 is an explanatory diagram showing an operating state of the switch of FIG.

9 is a waveform diagram showing the output voltage of the circuit of FIG.

FIG. 10 is an explanatory diagram showing the relationship between load voltage and load current.

[Explanation of symbols]

 1 charging circuit (adding means) 2 discharging circuit (subtraction means) 3 sensor Q2 comparator Q3 OR gate

Claims (1)

[Claims] 1. In a driving method of a switching power supply whose output is controlled by pulse width modulation, adding means for adding a predetermined ON time width to the pulse width of a switching pulse given from the outside, and the added pulse. A driving method of a switching power supply, comprising: a subtraction unit that subtracts an ON time width corresponding to a load current from the width, and switching control is performed by the subtracted pulse.