TWI275919B - Quick-recovery low dropout linear regulator - Google Patents

Abstract

This invention relates to a quick-recovery low dropout linear regulator (LDO), which utilizes a current-detection circuit to detect the magnitude of the output current, and the output current compares with a reference current so as to dynamically adjust the bias-current of the 2nd stage amplifier such that the system remains stable even when the output current is high resulting from the damping ratio zeta is still greater than 1. As a result, the output voltage can quickly recover stability from a large and sudden change of the output current.

Description

1275919 IX. INSTRUCTIONS DESCRIPTION OF THE INVENTION [Technical Field] The present invention relates to a low-dropout linear regulator (1 〇w dr op-out linear regula1: 〇r; LDO), and particularly relates to an output current suddenly large When changing, it can quickly return to a low-dropout linear regulator with a stable voltage output. [Prior Art] As the communication market matures, the application of related 1C is not growing. However, with the development of portable products such as mobile phones, the length of use of batteries is even more important. How to improve the power efficiency of the battery and maintain its stability is a very challenging topic. In recent years, low-drop P-out linear regulator (LDO) has become a low-power buck and regulator circuit because of its improved conversion efficiency and its small size and low noise. Mainstream. It is used in a large number of portable systems and communication-related electronic products that are powered by batteries. In the existing products (methods), in order to make the low-dropout linear regulator more accurate, a three-stage series operation amplifier is generally used to increase its gain, but it will cause instability. The situation. Therefore, various frequency compensation methods have been proposed to achieve system stability. At the beginning, it was proposed to increase the phase margin by using an external capacitor to lower the main pole position. However, this also has the following disadvantages: Since the main pole of such a circuit falls on the output point, it requires a large negative 1275919 load capacitance to stabilize the system. However, it is not easy to make this capacitor inside the chip, which increases system integration. Difficulties. In general, we want a larger system gain to improve the accuracy of the system. However, increasing the gain relative to the system reduces the stability of the system, thus causing a trade-off between accuracy and stability. The magnitude of the output current is limited by the stability of the system. Because the larger the output current, the smaller the load resistance, the larger the main pole at the output point, and the worse the stability of the system. Since the non-primary pole of this circuit is located at the output point of the operational amplifier (generally high impedance), the relative non-dominant pole of the first pole is lower, thus reducing the bandwidth of the system and also making it temporarily The state response is worse. Therefore, in order to improve the above shortcomings, various kinds of frequency compensation methods have been proposed, such as: nested Miller compensation, damping coefficient (control

System:··etc. However, these methods require two compensation capacitors, so (4) 'the size of the wafer is larger than the simple Miller compensation method, and the household uses a single Miller compensation using gain amplification, as shown in Figure 1 . Although this method successfully solves the above problem, the damping coefficient f is affected by the output current.

The speed of Vout敎. For the output dust, please refer to "Figure 2", which is "the first model" where gml, gm2, gmp are respectively. Two" ring number guide, gol £〇2 d A Chu, first and output stage Pass, '-,, and, second and round output conductance, 6.1275919

Cpi and Cg are the input parasitic capacitances of the second and output stages, respectively. C〇 is the negative load. Each Re is the parasitic resistance of the load cell, and cmi and Rm are the compensation capacitor and the compensation resistor. Adc is the DC gain of the system and system, and is the damping coefficient. The system model conversion equation can be derived from the small signal model (smaU signal) and "丨".

Av (s) ^λ 0 + sc out re^\ sc -)-s c mlc ^ m 2 S m p sc

Society At = · with p _ ^mlSmlSm r-3db ~ S〇\S〇2Sl a

R S〇2 SmlSi mP J' I SmlSt mp _ f〇f small lb2 a

In order to reduce the cost of Smr, the current market tends to

: The capacitor should be used as the load capacitor. However, the ceramic capacitor itself 2 = ruth is small, so the equation (3) can be simplified to (3). Ratio and:] The damping coefficient 'is inversely proportional to the output stage conduction (6), and the bias current (Ib2) of the amplifier, since the output power, - becomes large (for example, by 〇.lm "> 15 mA") , the output will become larger) 'So it will make the damping become smaller and ^ nearby ^ ΘΓ, υ 钱 'money frequency (10) should have a large wave near the unity gain frequency 'so that when the output current suddenly changes, the output voltage 7 1275919: The transient response of Vout will generate a continuous wave to slow down its stability, and it cannot provide a fast and stable low-dropout linear regulator. [Summary content] The main purpose of the present invention is to dynamically adjust the first The bias current of the secondary amplifier compensates for the influence of the output current variation on the damping coefficient , to eliminate the frequency response glitch and accelerate the stable speed of the output voltage. The present invention is a fast recovery low dropout linear regulator, It has a low-dropout linear regulator® circuit with a gain-amplified single Miller compensation capacitor, the low-dropout linear regulator circuit has a second-stage amplifier and has an output current a load is used, comprising a current detecting circuit, a comparing circuit, a control component and a pressurizing circuit, wherein the current detecting circuit is configured to detect the magnitude of the output current of the compensation circuit to output a corresponding one Comparing current; the comparison circuit inputs the comparison current and a fixed reference current to compare the size of the comparison circuit and generate a comparison signal; the control component controls and outputs a control signal by the comparison signal The pressing circuit 'to change the bias current of the second stage amplifier, the pressing circuit receiving the control signal of the control element to determine whether to change the bias current of the second stage amplifier; The comparison current is greater than the reference current, and the comparison circuit outputs the comparison signal to control the pressure circuit to increase the bias current of the second stage amplifier, thereby offsetting the influence of the output current variation on the damping coefficient 以 to reduce the transient response. The output voltage is connected to the wave and accelerates its steady speed. [Embodiment] 8 1275919 Details and technical description of the present invention Now, the following diagram is described as follows: Please refer to "Figure 3", which is a circuit diagram of the system of the present invention, which comprises a low-dropout linear regulator circuit 1 with a gain-amplified single Miller compensation capacitor (Miuer capacitor) That is, it supplies a required driving voltage by using a high-level supply voltage Vdd having a second-stage amplifier 1〇1 and having an output current for use by a load of φ, which utilizes The current mirror principle uses two transistors MP, MPR to form a current detecting circuit 2〇, and can use the amplifier 30 to form a negative feedback mechanism to accurately detect low-dropout linear voltage regulator. It is the output current 11G of circuit 1 and outputs a corresponding comparison current I2. And in order to reduce the load and consumption of the circuit, the comparison is Sil 12. The size value can be reduced by a multiple, such as let 12. = I iq / k ; where κ is determined by the characteristics of the transistors MP and MPR. The comparison current is compared with a fixed reference current Iref by a ratio Φ compared to the circuit 40 to output a comparison signal, the reference current Iref being generated by the high level supply voltages Vdd, Vb and the transistor MI, the comparison circuit The 40 series can be composed of a 1 to 1 NM0S current mirror 41, which respectively inputs the comparison current 12 〇 and the reference current Iref, and has a high impedance point Vr, and the two ends of the 咼 impedance point Vr are respectively the reference current I w and 1 to 1 0 0S current mirror 41 according to the comparison current generated by the same current 14 〇, by measuring the current change of the high impedance point Vr, can be used for current comparison, and generate a comparison signal. 9 1275919: The comparison signal is used for inputting to a control component 50 to generate a control signal, and the control component 50 can be constructed by using two series inverters invn, invp, and connected to the high impedance point Vr. The comparison signal generated by the current change of the high-impedance point Vr is generated by an inverter (invn, invp) and a control signal via the two series inverters invn and invp (Inverter). The control signal is used to control the pressurization circuit 6A, and the pressurization circuit Φ 60 is used to change the bias current of the second stage amplifier 101. The pressurization circuit 60 can be composed of transistors MSP, M24a, M22a and The MSN is configured to increase the bias current of the second stage amplifier 101 by controlling the conduction of the transistors MSP, MSN by the control signal. Before the operation of the invention, the magnitude of the reference current Iref must be determined first. The optimal design value of the reference current Iref is the damping coefficient of the system (the output current of the system is less than 1 守, if the output current is When the size of L is A, the damping coefficient of the system is less than i, then the value of the reference Φ current Iref can be set to A/K. This month can be divided into two states when operating, first Output current I!. At light load output, 'the damping coefficient of the system at this time is less than the reference current IKW' is less than the reference current Ire". At this time, the control signal of the inverter of the control unit 5〇2 series is Unvn: 〇, _ : n, let the switch of the MSP, MSN switch be open, so the bias current of the second stage amplification does not become large, then the damping coefficient of the system Γ still does not need to compensate ' Let the output current I1Q output voltage • 1275919

Vout is fast and stable. Large = current L. When changing to the heavy-duty output, the output coefficient of the output current ί10 becomes smaller, and the damping coefficient starts to become smaller. (4) The current starts to be unstable and is generated;

Crying: the fresh room of the machine ref, at this time, the control signal 50 in the reverse direction of the 11 in vp control signal is converted into (in vn : 1, in Vp : 〇), and the switches of the transistors MSP, MSN are turned on. By increasing the bias current of the second stage amplification crying, and because the damping coefficient 成 is proportional to the dust current of the second stage amplifier ^, the present invention increases the bias current of the second stage amplifier ι〇ι, The output current can be compensated [. For the influence of the damping coefficient Γ, let the damping coefficient Γ be simple (four) value (greater than υ. Therefore, when the output 曰 current L·. is converted from light load (such as 〇.lmA) to heavy load (such as 15 mA), The output voltage v〇ut of the output current I1G can also quickly reach a stable state. Please refer to "Fig. 4" and "Fig. 5", respectively, which are obtained when the low-dropout linear regulator circuit 10 is not compensated. And the transient response diagram. As shown in Figure 4, it is obvious that when the 曰 wheel: in is changed to the heavy load output, the frequency response diagram will have a glitch near the unit uni frequency (uni邙gain frequency), and therefore There will be a lot of chopping on the transient response graph of "Fig. 4", so that the stabilization time of the output voltage v〇ut takes 34 microseconds. Please refer to "6th" and "7th". The frequency response diagram and the transient response (transient 11 • 127591-9 response) obtained by the low-dropout linear regulator circuit 10 are shown in Fig. 6, which can be seen from Fig. 6. There is no surge near the unity gain frequency, so it can be seen in Figure 7. The chopping on the transient response is reduced so that it can be quickly stabilized (about 8 microseconds); therefore, the present invention can quickly output the output voltage V〇 when the output current L is changed to the heavy load output. The ut is restored to a stable state. The above is only a preferred embodiment of the present invention, and is not intended to limit the scope of the practice of the present invention, that is, all changes and modifications of the Φ according to the scope of the present invention are The invention is covered by the scope of the invention. [Simplified description of the drawings] Fig. 1 is a conventional low-dropout linear regulator. Fig. 2 is a small signal model of Fig. 1. Fig. 3 is a circuit diagram of the system of the present invention. Fig. 4 is a frequency response diagram of the present invention when uncompensated. Fig. 5 is a transient response diagram of the present invention when uncompensated. Fig. 6 is a diagram of the frequency response after compensation according to the present invention. , is the transient response diagram of the invention after compensation. [Main component symbol description] Ιι〇: output current I 20 · than the parent current 140 : current invn, invp: reverser

Iref: Reference current MP, MPR, MI: Transistor 12 1275919 MSP, MSN, M24a, M22a: Transistor Vdd, Vb: High level supply voltage Vr: High impedance point Vout: Output voltage 10: Low dropout linear regulator Circuit 101: second stage amplifier 20: current detecting circuit 30: amplifier • 40: comparison circuit 41: current mirror 50: control element 60: pressurizing circuit

13

Claims (1)

1275919 X. Patent application scope: 1. A fast-recovery low-dropout linear regulator with a low-dropout linear regulator circuit with amplifying and amplifying a single Miller compensation capacitor. The low-dropout linear regulator circuit has a second stage amplifier having a round current for use by a load, characterized in that it comprises: a current detecting circuit for detecting the output current of the low drop linear regulator circuit The size is to output a comparison current, Luyi comparison circuit, input the comparison current and a fixed reference current to use the comparison circuit for size comparison and generate a comparison signal; a control component, the control component is compared by the comparison The signal is controlled and outputs a control signal; a pressing circuit is used to change the bias current of the second stage amplifier, and the 5th pressing circuit receives the control signal of the control element to determine whether to change the second stage amplifier a bias current; accordingly, when the comparison current is greater than the reference current, the comparison circuit outputs the comparison signal to control the addition The second-stage circuit increases the bias current of the amplifier. 2. The low-dropout linear regulator according to claim 1, wherein the current detecting circuit is formed by a current mirror circuit. 3. The low-dropout linear regulator of claim 2, further comprising an amplifier that forms a negative feedback mechanism with the current detecting circuit. 4. The low-dropout linear regulator as described in claim 1, 14 1275919, wherein the comparison circuit is formed by a 1 to 1 NMOS current mirror, and the comparison circuit inputs the comparison current and the reference current, respectively, and The comparison circuit has a high-impedance point, and the two ends of the high-impedance point are respectively a current of the reference current and the current mirror of 1 to 1 NM0S according to the current generated by the comparison current, by measuring the current change of the high-impedance point, It can be used for current comparison. 5. The low-dropout linear regulator according to claim 1, wherein the control element is formed by connecting two inverters in series, so that the control signal has (0, 1 signal output of (invn, invp)). . 6. The low-dropout linear regulator of claim 1, wherein the pressurizing circuit is formed by a transistor, and the control signal is used to control whether the transistor is turned on or off to increase the second stage. The bias current of the amplifier. 7. The low dropout linear regulator of claim 1, wherein the magnitude of the comparison current is a multiple of the output current. 15