KR800000846B1 - Control circuit arrangemant for generating control signal for a voltage converter - Google Patents

Abstract

A control circuit arrangement generates a periodic pulsatory control signal for controlling a switch in a converter for converting an input direct voltage into an output direct voltage which is substantially independent of vibrations of the input voltage. The circuit comprises a current source, a controllable switch(s), a capacitor, a means connecting the current source and controllable switch to the capacitor so as to generate a sawtooth voltage across the capacitor and a threshold level detector(Dr) for converting the sawtooth voltage into the periodic pulsatory control signal.

Description

Control signal generator for voltage conversion

1 is a first principle diagram of a circuit diagram of the circuit adjustment apparatus of the present invention.

2 is a waveform diagram of FIG.

3 is a diagram of various modified waveforms in FIG.

4 is a schematic diagram of a circuit control device according to the waveform of FIG.

5 is a part of a circuit portion.

6 is a circuit diagram secondary principle diagram of a circuit control device.

7 is a waveform diagram in FIG.

8 is a third principle diagram of a circuit diagram of a circuit controller.

9 is a partial view of a circuit diagram in which the circuit control device of the present invention will be installed in a television receiver.

10 is a circuit regulation diagram called a so-called up converter.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an adjustment circuit for generating a periodic pulse current adjustment signal for adjusting a switch of a converter for converting an input DC voltage into an output DC voltage independent of the input voltage. The adjustment circuit includes a current source and a capacitor. An adjustable switch for generating a sawtooth voltage, and a detector for converting the sawtooth voltage into a pulse control signal, the pulse width is adjustable by adjusting the voltage.

Negative feedback is widely used in these circuits. The voltage fed back from the output voltage is compared with the reference voltage, and the output voltage is determined according to the error signal formed in this way. The pulse width of the adjustment signal changes. As a result, the output voltage is independent of the change in the input voltage caused by the rectifying action.

If the amplification degree of the negative feedback circuit is large enough, the effect of the slow change of the input voltage on the output voltage can be eliminated. However, the effects of rapid change are difficult to eliminate. The loop itself causes delay because the output voltage appears at the capacitor and the given lease width is related to the stabilization factor. The rapid change in input voltage is due to the ripple remaining after rectification. Normally, the adjustment signal cannot eliminate the leading eadge of the ripple voltage. For example, in a television receiver, the deflection vtg must be kept constant within 0.1% so as not to disturb the change of the video signal width. When ripple voltage is 10% of input voltage, stability should be 100.

The solution is to make the ripple voltage more gentle. This can be solved by using a large capacitance electrolytic capacitor choke coil or the like. This is expensive and bulky. Another drawback remains. In other words, as the input voltage value increases, the loop amplification degree also increases. However, in order to avoid such amplification instability, the input voltage should be lowered to an appropriate value.

French patent application 2,225,879 published a control circuit of a converter connected to a forward control with a feedback coupling. Information about the change in the input voltage is transmitted to the control circuit so that the conduction time of the converter switch is affected. By the selection of the information, it is possible to carry out the whole compensation which was only partial for other patents.

It is an object of the present invention to provide a regulating circuit which can essentially eliminate all the effects of a change in the input voltage, which has no effect on the ripple voltage, and which can be used in any type of converter. Therefore, the characteristics of the adjustment circuit of the present invention are as follows. First and second circuits for adjusting the current supplied from the current source by the linear function of the input DC voltage and the control voltage, and controlling the change of the sawtooth voltage during the generation time of the control signal with the input DC voltage. In order to adjust the voltage as a linear function, a second circuit having first and second input terminals, two first input terminals that can be connected to an input DC voltage, and two second input terminals that can be connected to a regulated voltage Etc.

The output voltage to be measured depends only on the regulated voltage and can be adjusted freely and measured accurately. The circuit according to the invention is characterized in that it uses a feedback circuit to supply a regulated voltage. This feedback circuit includes a comparison stage for comparing the output voltage with the reference voltage. As a result, the circuit has both the benefits of forward regulation and the benefits of known backwards.

The present invention will be described in detail with reference to the drawings.

1 shows a circuit for supplying a voltage by a switch action in series. The voltages appearing at the two input terminals (1) (2) are rectified by the rectifier (3) and rippled by the capacitor (4). The DC voltage V _B appears in the capacitor 4 and depends on the change of the input voltage. The other terminal of the capacitor, which is not connected to the rectifier 3, is grounded. The converter comprises an npn transistor, a tapped inductance L with a cathode connected to the diode and a capacitor 5. The collector of the transistor is connected to the junction of the elements 3 and 4, and the emitter is connected to the inductance L. The anode and the capacitor 5 of the diode D are grounded. The remaining terminal of the inductance L shows the output voltage V ₀ . The load 7 is connected between the terminal 6 and the ground.

Periodic conditioning pulses are applied to the base of the transistor to cause the transistor to alternately conduct and fail.

If the time δT at which the transistor is conducting is part of the period T, and the inductance is 1: n between the whole and the tap and the terminal 6, then V _B and V ₀ become the following equation.

This relationship indicates that the inductance L is an incomplete discharge at the transistor's dead time.

In the known circuit, δ can be changed by negative feedback and pulse width modulation. In that way, the change in output voltage V ₀ becomes independent of the change in input voltage V _B. There are special cases where V ₀ must be kept constant. In this case V ₀ must be compared to the reference voltage.

The regulating circuit of the transistor in FIG. 1 comprises an oscillator 8 which generates a pulse of frequency 10-20 KHz. This pulse is applied to an adjustable switch S in series with the voltage source V ₁ . The capacitor C and the current source I are connected in parallel with the series arrangement described above, and the oscillators 8, V ₁ , C, and I are grounded. Capacitor C is connected to one input terminal of the threshold level detector Dr, the other input terminal is connected to voltage source V _2, and voltage V ₂ is lower than voltage V ₁ . The output of the detector Dr adjusts the transistor base and thus the drive stage.

2 shows that the voltage Vc of capacitor C is a function of time. Due to the short conduction time of the switch S, the voltage Vc has a value of V1 so that the transistor is turned off. When switch S is off, capacitor C is discharged by current I. Therefore, the slope of the sawtooth wave is determined by the current I because the voltage Vc decreases linearly. As soon as V ₁ falls to V ₂ , the transistor is conductive. It is still conducting after the time interval δT until the switch S is energized for a short conduction time appearing at the beginning of a new period.

Where K is a constant and Vr is a regulated voltage.

With these values, current I reduces the capacitor's voltage to the next level after one cycle.

Vc becomes (V ₂ ) after (1-δ) T to satisfy the following relation.

When charged to the above value, the regulated voltage V ₂ becomes

If you substitute equation (1) here,

This is the regulated voltage expressed as a function of the desired output voltage V ₀ . In other words, since the output voltage is independent of the change in the input voltage without using negative feedback, it is determined by the adjustment voltage. The regulated voltage can be precisely adjusted so that the output voltage is fixed and can supply more power.

=300V에 대해서 Vr=3V가 된다. 캐패시터 C=4.7nF에 대해서, 전류 I는 두 전류의 합이다.In a practical circuit of a television receiver, a line oscillator is used as the oscillator 8. The oscillator supplies a pulse of T = 64 / kHz. If n = 0.8, K = 0.01,

Vr = 3V for = 300V. For capacitor C = 4.7nF, current I is the sum of the two currents.

와

이다. 만약 제1 전류가 V_B로부터 유입되면 저항 R_B=

=1.36MΩ이 된다.In other words,

Wow

to be. If the first current flows from V _B , the resistance R _B =

= 1.36MΩ.

=68KΩ이 된다.Similarly, the second current flows from V ₂ so that the resistance RT =

= 68 KΩ.

And, V ₁ and V ₂ are variable resistors and can be derived from V _B and V ₂ .

In FIG. 2, capacitor C is completely discharged such that Vc becomes zero before the end of the cycle. Of course this is not necessary. The dotted line in the figure shows the change in Vc when the above-mentioned voltage decreases linearly during the entire period. It should be noted that the level at which the transistor can conduct is greater than V ₂ increased by the second current.

That is, V ₂ + (1-n) Vr = Vr, which is the regulating voltage, and the change of the voltage Vc during the period (1-δ) T of the transistor is ΔV = V ₁ -V ₂ = KV _B -nVr. From this, it can be seen that I and ΔV are linear functions of V _B and Vr. In the special case of n = 1, the diode D is not connected to the tap of the inductance L, but to the junction with the emitter of the transistor, and the current I is independent of Vr. When ΔV changes, the value of Vc does not change at the last moment of the period T because V _B changes. In fact, this last value depends on terminals V ₂ and Vr.

Above, it is assumed that the conduction time of the switch S is infinitely short and coincides with the moment when the cycle is completely finished. In practice, however, the transistor has a finite time of 7-10 milliseconds and the switch S must be energized for at least the time ts described above. The switch S can be made longer.

3 shows the change in Vc after the switch S is turned on for ts, ㎲ before the end of the cycle, and turned off for βT㎲ before the end of the next cycle. From FIG. 3, the magnitude of the sawtooth voltage KV _B + (1-n) V ₂ should be multiplied by β, and the sawtooth shape is cut off.

The embodiment of the circuit according to the invention is shown in FIG. The line oscillator 8 applies a pulse of width ts + (1-β) T to the base of the transistor 9, and the collector adjusts the base of the switch S composed of emitter followers. Then, the switch is turned on while the pulse of the oscillator 8 is generated. Put resistor 10 between transistor S and K _B and resistor 11 between base and Kr and connect resistor 12 and diode 13 in series between base and ground, and K _B is connected to V _B. By connecting K ₂ to Kr, the emitter of the transistor S can have a voltage β [KV _B + (1-n) Vr] during the conduction time. For example, when K = 0.01 β = 0.75 n = 0.5, the values of the resistors 10, 11 and 12 are 390 KΩ, 7.8 KΩ and 4.7 KΩ. The voltage drop of the diode 13 is compensated by the voltage difference between the emitter and the base of the transistor 5.

Capacitor C is connected between the emitter of transistor S and transistor 14, and is connected between the base and resistor Rr. RB ₁ and the adjustable RB ₂ are connected in series.

The resistors Rr and RB ₂ are connected to the terminal terminals of Kr and K ₈ , K _B is connected to V _B, and Kr is connected to Vr. RB ₂ is adjusted such that the sum of RB ₁ and RB ₂ is equal to the value of RB. The transistor 14 and the emitter are grounded, and the collector of the transistor 14 and the collector of the transistor S are connected to each other through a resistor 15 of 1.5 KΩ. During the charge time of capacitor C, resistor 15 limits the collector current of transistor 14.

During the time interval βT, the transistor 9 is turned on so that the transistor 6 is turned off. Capacitor C discharges through resistor 15, and discharge current I becomes the collector current of transistor 14. As shown in FIG. Therefore, it is much larger than the base current that causes the transistor to conduct. Devices S, 14, 16 and C form Miller OSC such that the voltage Vc appearing at the emitter of transistor S has good linearity. Since capacitor C is at the base of the transistor, Rr and R _B1 + R _B2 can be regarded as current sources. The threshold level detector Dr consists of a pnp transistor, the emitter has a voltage Vr, and the base is connected to a voltage Vc. During the time interval δT, the transistor is turned on, and a positive pulse is formed in the collector. This pulse is inverted by transistor 16 and drive transistor 17 is turned off. The transistor is regulated by the converter and conducts for δT, and transistor 17 is turned off. The voltage Vr is represented at point A by a zener diode 18 in series with the complementary diode.

The circuit shown in FIG. 4 includes a first circuit having terminals K _B and Kr and a second circuit having K _B 'and Kr'. The electronic terminals are respectively connected to voltages V _B and Vr, and when the correct value is selected for the resistors 10, 11 and 13, the magnitude of the voltage Vc has a desired value. When terminal _B K 'and K _2' is connected to each of V _B and V _2, the exact value has been selected with respect to R _B + R _B2, the current I has a desired value. The magnitude and current described above are linear functions of voltages V _B and Vr. When n = 1, Kr and Kr 'are not connected.

The description above is directed to forward regulation. However, it may be desirable to use reverse adjustment. For this purpose, the regulated voltage can be determined by feedback coupling. The advantage of this is that the influence of tolerance and temperature is eliminated by regulation. This eliminates the influence of the change of the load 7. Since the voltage V ₀ is proportional to Vr, the amplification degree of the negative feedback is constant. Therefore, this loop can be fabricated without the risk of instability at high input voltages. 5 shows how the voltage Vr can be obtained. As the variable resistors 19 and 20, the voltage obtained from V ₀ is applied to the input terminal of the differential amplifier 21, and the reference voltage appearing at the control diode 22 is applied to the other input terminal of the amplifier 21. When the difference between the two input voltages of the amplifier is amplified to the desired value of the regulated voltage Vr, this voltage is used at the output of the amplifier. The current coming from the voltage source flows through the control diode 22 and this power source capable of supplying the collector current of transistor S in FIG. 4 can be obtained from V _B and V ₀ . Thus, the circuit of FIG. 5 can replace the zener diode 18 of FIG.

In a known circuit using only negative feedback coupling, it should be noted that δ is adjusted according to the output voltage V ₀ . In the device described above, δ varies with the input voltage V _{B and} is regulated under the influence of a change in voltage V ₀ .

The output voltage should increase slowly after conducting. Otherwise, the peak current through the transistor becomes too large because the capacitor 5 has not yet been charged. This can be done by allowing V _B to increase slowly, but it would be more practical to raise δ slowly from zero. As a result, however, the ripple voltage at the input is delivered to the output, which causes the peak current through the transistor to be too large. As a result, the stable circuit can react so that the supply circuit cannot be driven. The solution to this problem is to raise the regulating voltage Vr slowly during departure.

The embodiment is shown in FIG. The capacitor 23 is charged by the current coming from the power supply V _B flowing through the resistor 24, and the time constant becomes large. The junction of elements 23 and 24 is connected to the output of differential amplifier 21 via diodes 25 and 26. After the switch is turned on, the voltage slowly increases. As a result, δ and V ₀ also increase slowly. As a result of the operation of the amplifier 21, the voltage Vr has a high value. As a result, the diode 25 becomes conductive. The junction of diodes 25 and 26 is connected to point A in FIG. 4 so that the voltage acting as a regulating voltage at point A starts slowly. When the voltage at the junction of the resistors 19 and 20 reaches the reference voltage value of the zener diode 22, the voltage Vr decreases. At any moment diode 26 is turned on and diode 25 is turned off. As a result, V ₀ and Vr are proportional to each other.

[nV_B+(1-n)Vr〕여기서 △V는 트랜지스터의 불통 시간동안 전압 Vc의 변화이다. 그러면 도통시간의 변화는 Vr과 같고,What is described above relates to a circuit in which the transistor is also conducted at the end of the cycle. The circuit of Fig. 1 and its embodiment is such a method in which the transistor is passed even at the beginning of the discharge time of the capacitor C. In FIG. 2, the input delta of the sensor Dr must be interchanged with the time δT and (1-δ) T and in FIG. If V = n (V _B -Vr), I =

[nV _B + (1-n) Vr] where ΔV is the change in voltage Vc during the discontinuation time of the transistor. Then the change in conduction time is equal to Vr,

=nδV_B+(1-n)δV₂이며, 식(1)과는 다르다. 이 경우 I와 △V는 V_B와 Vr의 선형함수이며, V₀와 Vr은 서로 비례한다.Vr =

= nδV _B + (1-n) δV ₂ , which is different from equation (1). In this case, I and ΔV are linear functions of V _B and Vr, and V ₀ and Vr are proportional to each other.

In FIG. 6, the capacitor C is not discharged by the current I as in FIG. 1, but in the case of charging, it is the same as in FIG. No voltage source in series with switch S is required. In this figure, only important elements are shown.

In FIG. 7, the voltage Vc of the capacitor is plotted as a function of time. In FIG. 2, it is sawtooth descending while in FIG. 7 it is rising sawtooth. Since the voltage V _B changes when ΔV changes, the value of the voltage Vc in FIG. 7 does not change at the initial instant of the period T. If the transistor is conducting in the first part of the cycle, and has a decreasing sawtooth shape, it will satisfy Equation (1). Tweaked during the time of the I and the transistor, the change of the sawtooth voltage △ V is a linear function of V _B and Vr. If the transistor conducts in the second part of the increasing sawtooth period, this can be demonstrated as in the case of the decreasing sawtooth wave.

All embodiments relate to a transducer in series to which equation (1) can be applied. FIG. 8 compares with FIG. 1 and shows a circuit for a parallel converter with inductance L and diode D switched. The voltage V _B must be negative, the transistor is of pnp type, and the circuit is regulated in the same manner as in FIG. For serial converters and parallel converters, we can prove that the following relation applies.

Formula (2) is the same as Formula (1) when m = n.

The parallel converter satisfies equation (2) when m = 0. In view of the similarity between equations (2) and (1), it is clear that the circuit according to the invention can be used in the regulator of a parallel converter. In addition, the current I and the voltage change ΔV have a similar proportional relationship.

According to the circuit arrangement in the formula (2) or according to the circuit arrangement of the present invention, the supply voltage circuit and the horizontal deflection circuit of the television receiver are used in combination, which is shown in FIG. Where Ly is a horizontal deflection coil, Ct is a trace capacitor, Cr is a retrace capacitor, D ₁ is a parallel diode, inductance L is transformer T ₁ , and is not connected to the tap of winding 27 of transformer T ₂ . The transformer T ₁ has a transformer ratio of 1: n, and the ratio of turns of the entire winding to the taps 27 is 1: m, where n and m are variables shown in equation (2). The voltage supplied to the part of the receiver and the high tension at the end of the screen are formed in the secondary winding of transformer T ₂ . Transformers T ₁ and T ₂ have a core.

The voltage appearing on the capacitor 5 in series with the winding 27 serves as the output voltage V ₀ . As a method of the circuit arrangement of FIG. 5, the voltage Vr is obtained by the amplifier 21 which is an npn transistor. A low resistance 28 is applied between the diode 25 and the junction of the resistor 24 and the capacitor 23, and the thyristor 29 is connected between ground and the junction of the diode 25 and the resistor 28. Thyristor 29 is conducted by safety circuit 29 'when the current consumed by the circuit is large enough to cause capacitor 23 to discharge. The junction of diodes 25 and 26 is connected to point A via an emitter follower. The voltage across zener diode 30 connected to the emitter terminal of transistor 17 serves as a stable supply voltage for the starting circuit, switch S and oscillator 8, which appears immediately after the switch is turned on. .

In fact, it is chosen as:

n = 0.49

m = 0.29

β = 0.8

k = 0.01

C = 4.7 nF

The value of the resistor 10 is 390 KHz, and the oscillator 8 uses an integrated circuit such as a Philips type TBA 920. It has a constant value of 140V, the voltage V ₀ is stable, the change in voltage V _B is 200-370V at the main frequency, there is no ripple. The capacitor 4 is of a smaller capacity than 100 mW.

Figure 10 shows a switched transducer in which equation (II) does not apply. In this circuit, the inductance L is connected between V _B and the switch Tr, and the diode D is connected between the tap of the inductance L and the output terminal 6 ^o . If the inductance ratio is 1: n, the following relationship is established.

Here, it can be seen that the circuit arrangement of the present invention is used when the sawtooth wave voltage increases or decreases. 10 shows a detector Dr. FIG. Here again, the change in current I and voltage ΔV is a linear function of the control voltage Vr proportional to the input voltage V _B and the desired output voltage V ₀ . ΔV is the change in voltage Vc during the time δT when the transistor is turned on. In the special case of n = 1, where diode D is not connected to the tap of inductance L but to the junction of the transistor's collector, terminal K _B ′ is not connected to VB because current I is independent of voltage V _B.

This supply voltage circuit can be used to generate a constant output voltage. One application is to cause voltage V ₀ to change with changes in regulated voltage. Such a case appears in a color television receiver, and if the regulating voltage Vr changes with frequency and parabolic function, the voltage V ₀ becomes the supply voltage of the line deflection circuit. As a result, the line deflection current is modulated at the frequency required for east-west-corection.

Claims (1)

It consists of a current source, an adjustable switch (S), a threshold level detector (Dr), a first circuit with two input stages, and a second circuit to convert the input voltage into an output voltage independent of the change in the input voltage. An adjusting circuit device for generating a periodic pulse adjusting signal for adjusting a switch.

Images