TW200845546A - Low dropout (LDO) linear voltage regulator - Google Patents

Low dropout (LDO) linear voltage regulator Download PDF

Info

Publication number
TW200845546A
TW200845546A TW96115423A TW96115423A TW200845546A TW 200845546 A TW200845546 A TW 200845546A TW 96115423 A TW96115423 A TW 96115423A TW 96115423 A TW96115423 A TW 96115423A TW 200845546 A TW200845546 A TW 200845546A
Authority
TW
Taiwan
Prior art keywords
capacitor
power transistor
output
miller
voltage regulator
Prior art date
Application number
TW96115423A
Other languages
Chinese (zh)
Other versions
TWI329967B (en
Inventor
Wei-Ren Huang
Original Assignee
Sitronix Technology Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Sitronix Technology Corp filed Critical Sitronix Technology Corp
Priority to TW96115423A priority Critical patent/TW200845546A/en
Publication of TW200845546A publication Critical patent/TW200845546A/en
Application granted granted Critical
Publication of TWI329967B publication Critical patent/TWI329967B/zh

Links

Landscapes

  • Amplifiers (AREA)
  • Continuous-Control Power Sources That Use Transistors (AREA)

Abstract

A low dropout (LDO) linear voltage regulator is provided to employ a nested Miller compensation (NMC), is referred to as the basic structure, for designing a low dropout voltage regulator (LDO) without an output capacitance. An active resistor may be added to the Miller capacitance feedback circuit. Thus, the increase of the damping coefficient control can accommodate the wide range of the output capacitance and its parasitic resistance, and further resolve the tradeoff between the damping coefficient control and the system loop gain. Furthermore, the use of capacitor-sharing switch technique reduces the Miller capacitance value in the whole system, and thereby reduces the output-voltage stabilization time without affecting its stability.

Description

200845546 九、發明說明: 【發明所屬之技術領域】 一種低壓降線性電壓穩壓器’由指一種使用主動電阻 與分享電容切換技術,而可以不需輸出電容的低壓降線性 電壓穩壓器。 【先前技術】 在通訊市場逐漸成熟發展之際,相關ic的應用更是不 斷成長,隨著如手機等可攜式電子產品的發展,供應電源 春的電池使用時間的長短更顯的額外重要。如何提高電池的 效率並且維持其一定的穩定度是一個相當有挑戰的課題。 近年來低壓降線性穩壓器因為其轉換效率的提昇,加 上其小體積、低雜訊的特性,成為小功率降壓與穩壓電路 的主流。在各式由電池供應電源的可攜式系統以及通訊相 關的電子產品上,皆被大量地使用。 在現有的產品(方法)中,為了讓低壓降線性穩壓器 更加地精確,一般而言會採用三級放大器來增大其增益, 然而卻會因此造成不穩定的情形。所以,不斷地有人提出 * 各式各樣的頻率補償方法來達到系統的穩定。一開始,有 人提出利用外加大電容來降低主極點位置而增加相位邊 際(phase margin )。即極點-零點補償(pole-zero compensation)的方法,因此需要大尺寸的輸出電容將主 極點推向低頻以維持穩定度,但是,由於其主極點位於輸 出點,最大負載電流值會影響其穩定度而有所限制。此卻 也有如以下之缺點: 1·由於此類電路的主極點落於輸出點,因此需要較大 輸出電容來穩定系統,然而實際產品並不容易將此電容做 5 200845546 • 在晶片内部,如此會增加系統整合性上的困難。 2· —般而言,我們希望較大的系統增益來提升系統的 精準度’然而此類電路提升增益,相對的卻降低了系統的 穩定度,因此會造成精準度與穩定度之間的取捨。 3·此類電路負載電流大小會受限於系統的穩定性。因 為輸出電流越大,代表著負載電阻越小,因此相對的位於 輸出點的主極點也越大,因此會使的系統的穩定性越差。 因此,為了改善以上的缺點,有人提出了使用巢式米 勒補償的低壓降線性電壓穩壓器(如「第1圖」所示),該 低壓降線性電壓穩壓器1〇包括一輸入端VIN接收輸入的直 流電壓,並有一輸出端VOUT輸出一穩定的輸出電壓;一功 率電晶體13及串聯的一第一級放大器11與一第二級放大 器12;功率電晶體13的源極與輸入端VIN相連,其汲極與輸 出端VOUT相連接,其閘極與第二級放大器12的輸出端相 連。第一級放大器11的反相輸入端由一參考電壓產生器14 輸入一參考電壓訊號,其同相輸入端連接炱一節點15 ’且 其輸出端連接至該第二級放大器12的輸入端’且輸出端與 # 功率電晶體13的汲極間的回授路徑上設有一第一米勒補 償電容Cml,而且第二級放大器12的輸出端與功率電晶體 13的没極間的回授路徑上設有一第二米勒補償電容Cm2 ° 一回授網路20係連接在功率電晶體13的汲極與第一級放 大器11的同相輸入端之間,該回授網路20係包括電阻rfi 和RF2,該兩電阻RF1、RFj·成了一分壓器,且兩電阻rfi、 RF2間構成該節點15,第一級放大器11的同相輸入端連接 至該節點15。該低壓降線性電壓穩壓器10的輸出端ν〇υτ並 外接有一輸出電容CL與其寄生電阻Resr。 200845546 該巢式米勒補償的低壓降線性電壓穩壓器〗〇係利用 極點分列(pole splitting)的特性讓主極點改變至第一級 放大器11的輪出點,經由這個方法可以不需要大尺寸輸出 電容CL的使用’甚至零輸出電容Cl也可以使得系統有很好 的穩定度4此有利於祕單晶片(s〇c)的制,並可 減少外部元件的使用以及縮小電路板的面積。200845546 IX. Description of the Invention: [Technical Field of the Invention] A low-dropout linear voltage regulator is a low-dropout linear voltage regulator that uses active resistors and shared capacitor switching techniques without the need for output capacitors. [Prior Art] As the communication market matures, the application of related ic continues to grow. With the development of portable electronic products such as mobile phones, the length of battery life used for power supply springs is even more important. How to improve the efficiency of the battery and maintain its stability is a very challenging topic. In recent years, low-dropout linear regulators have become the mainstream of low-power step-down and voltage regulator circuits because of their improved conversion efficiency and their small size and low noise. It is widely used in a variety of portable systems that supply power from batteries and communication-related electronic products. In existing products (methods), in order to make the low-dropout linear regulator more accurate, a three-stage amplifier is generally used to increase its gain, but it will cause instability. Therefore, it has been repeatedly proposed * a variety of frequency compensation methods to achieve system stability. In the beginning, it was proposed to use an external capacitor to lower the main pole position and increase the phase margin. That is, the pole-zero compensation method, so a large-sized output capacitor is required to push the main pole to the low frequency to maintain stability. However, since the main pole is at the output point, the maximum load current value will affect its stability. Degree is limited. This also has the following disadvantages: 1. Since the main pole of such a circuit falls on the output point, a larger output capacitor is needed to stabilize the system, but the actual product is not easy to do this 5 200845546 • Inside the chip, so Will increase the difficulty of system integration. 2. In general, we want a larger system gain to improve the accuracy of the system. However, such circuits increase the gain, but reduce the stability of the system, thus causing a trade-off between accuracy and stability. . 3. The magnitude of the load current of such circuits 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. Therefore, in order to improve the above disadvantages, a low-dropout linear voltage regulator (shown in FIG. 1) using a nested Miller compensation has been proposed, and the low-dropout linear voltage regulator 1〇 includes an input terminal. VIN receives the input DC voltage, and has an output terminal VOUT that outputs a stable output voltage; a power transistor 13 and a first stage amplifier 11 and a second stage amplifier 12 connected in series; the source and input of the power transistor 13 The terminal VIN is connected, the drain thereof is connected to the output terminal VOUT, and the gate thereof is connected to the output terminal of the second stage amplifier 12. The inverting input terminal of the first stage amplifier 11 is input with a reference voltage signal by a reference voltage generator 14, and its non-inverting input terminal is connected to the first node 15' and its output terminal is connected to the input terminal of the second stage amplifier 12'. A first Miller compensation capacitor Cml is disposed on the feedback path between the output terminal and the drain of the # power transistor 13, and the output terminal of the second stage amplifier 12 and the non-polar feedback path of the power transistor 13 are disposed on the feedback path. A second Miller compensation capacitor Cm2 is provided. A feedback network 20 is connected between the drain of the power transistor 13 and the non-inverting input of the first stage amplifier 11. The feedback network 20 includes a resistor rfi and In RF2, the two resistors RF1, RFj· become a voltage divider, and the node 15 is formed between the two resistors rfi and RF2, and the non-inverting input terminal of the first stage amplifier 11 is connected to the node 15. The output terminal ν 〇υ τ of the low-dropout linear voltage regulator 10 is externally connected with an output capacitor CL and its parasitic resistance Resr. 200845546 This nested Miller compensated low-dropout linear voltage regulator uses the characteristics of pole splitting to change the main pole to the turn-off point of the first-stage amplifier 11. This method does not require a large The use of size output capacitor CL 'even zero output capacitor Cl can also make the system have good stability. 4 This is beneficial to the system of single chip (s〇c), and can reduce the use of external components and reduce the area of the board. .

凊再參閱「第2圖」所示系統的小訊號模型,該小訊 ”包含了第一級放大㈣與第二級放大㈣的兩個 增显級gm]vs、gw%和一個功率電晶體13的輸出級,1中 Rfi、心2組成回授網路,gm〗、gw、分別為第一、^二 $ =、,輸出級的轉導’ ‘、R〇冰〜&分別 …弟、、及弟一級放大器11、12的輸出阻抗和寄生電容, 出電容,Resr為輸出電容的寄生電阻’〜、^ 為萆替t米勒補償電容,Rwt (=Rl|ir°pII(Rfi+Rfi)) 其中Rl>R〇P分別為負載阻抗和功率電 阳體的輸出阻抗。 式如^ :「第2圖」的小訊號模型可推導得其系統轉換方程 Λ凊Refer to the small signal model of the system shown in Figure 2, which contains two levels of gm]vs, gw% and a power transistor for the first stage amplification (four) and the second stage amplification (four). 13 output level, 1 in Rfi, heart 2 constitutes the feedback network, gm〗, gw, respectively, first, ^ two $ =,, the output level of the transduction ' ', R〇冰~& respectively... , and the output impedance and parasitic capacitance of the first stage amplifiers 11, 12, the output capacitance, Resr is the parasitic resistance of the output capacitor '~, ^ is the compensation capacitor for t Miller, Rwt (=Rl|ir°pII(Rfi+ Rfi)) where Rl>R〇P is the output impedance of the load impedance and the power galvanic body respectively. The small signal model of the formula: "Figure 2" can be derived from the system conversion equationΛ

c η ^ 'S —--^ __ mv^mJ mp ι+-c η ^ 'S —--^ __ mv^mJ mp ι+-

C〇utReSRC〇utReSR

1 + -^-V P~3dB l + s ^out^esr + Sm2 + s "mT^QUr SmlSmp ( (1) 其中直流增益為4 經 主極點為 p__3de C—mpRmR〇1R( 由推倒解出阻尼係數為 O2±V0UT f ζ' 21 + -^-VP~3dB l + s ^out^esr + Sm2 + s "mT^QUr SmlSmp ( (1) where the DC gain is 4 and the main pole is p__3de C-mpRmR〇1R (damped by the push-down The coefficient is O2±V0UT f ζ' 2

VV

Rf2Rf2

Rf2 ^OUT^ESR ~ SmlRf2 ^OUT^ESR ~ Sml

SmlSi mpSmlSi mp

Cm2CaCm2Ca

UT (2) (3) (4) 7 200845546 然而,由以上推導可知阻尼係數ζ會因為輸出電容值 以及其寄生電阻值resr的不同而改變,當輸出電容值c〇 以及其寄生電阻值Resr非常小時,為了得到一個足夠的阻 尼係數ζ和使用較小的米勒補償電容值Cm2,第二級的轉 導gm2必須縮小,如此系統回授增益將會相對變小而降低 了系統的精準度,形成阻尼係數控制ζ與系統迴路增益間 相互的取捨。 【發明内容】 • 爰是,本發明之主要目的係利用巢狀米勒補償為基本 架構,利用極點分列的特性讓主極點改變至第一級放大器 的輸出點,經由這個方法可以不需要大尺寸輸出電容的使 用,甚至零輸出電容也可以使得系統有很好的穩定度的優 點,k出在米勒電谷的回授路徑上加入主動電阻,藉此增 加阻尼係數的控制,以因應大範圍輸出電容與其寄生電阻 的使用,進一步解決阻尼係數控制與系統迴路增益間相互 的取捨。 # 本發明之再一目的係提出一分享電容切換技術來減 少整個系統所需的米勒電容值,如此可以減少整個系統所 需的米勒電容值,而且因為總電容值減少,可以進一步延 展頻寬並加速輸出電壓的穩定時間。 本發明係一種低壓降線性電壓穩壓器,其包括一輸入 端接收輸入的直流電壓,及一輪出端輸出一穩定的輸出電 壓;一功率電晶體,其源極與輪入端相連,其汲極與輸出 端相連接;一第一級放大器,其反相輸入端由一參考電壓 產生器輸入一參考電壓訊號,其同相輸入端連接至一節 8 200845546 點’其輪出端與該功率電晶體的汲極間設有一第一米勒補 償電容;一第二級放大器,其輸入端連接該第一級放大器 的輸出端,其輸出端與該功率電晶體的汲極間設有串聯的 一第二米勒補償電容與一主動電阻;以及一回授網路係 ,接在該功率電晶體的汲極與第一級放大器的同相輸入 &之間’该回授網路係由兩電阻構成的一分壓器,且兩電 阻間構成該節點。其中該主動電阻係連接成二極體形式的 電晶體。 本發明使用巢狀米勒補償為基本架構,利用極點分列 的特性讓主極點改變至第一級放大器的輸出點,經由這個 方法可以不需要大尺寸輸出電容的使用,甚至零輸出電容 也可以使得系統有很好的穩定度,如此有利於系統單晶片 的應用,並減少外部元件的使用以及縮小電路板的面積。 由於阻尼係數會因為輸出電容值以及其寄生電阻值的不 同而改變,阻尼係數不足時會導致頻率響應在單增益頻率 附近有突波產生,繼而使得輸出電壓對負載電流的步階響 2在準線性區(quasi-linear region)會有漣波產生而減緩 其%定的時間。因此本發明在米勒電容的回授路徑上加入 f動電阻,藉此增加阻尼係數的控制以因應大範圍輸出電 谷與其寄生電阻的使用,並可以進一步解決阻尼係數控制 與系統迴路增益間相互的取捨。 本發明之低壓降線性電壓穩壓器進一步包括一分享 :各電路,該分旱電容電路包含一分享電容,該分享電容 私路偵測該功率電晶體上的電流,且切換該分享電容分別 與該第一米勒補償電容或第二米勒補償電容產生並聯,藉 此減少整個系統所需的米勒電容值,並可以進一步加速輸 9 200845546 出電壓的穩定時間。 其中’鸪分享電容電路包括一電流感測電路,其用以 偵測該功率電晶體上的電流;一史密特觸發電路,其接入 =電流感測電路的訊號,並將訊號傳至一非重疊時脈產生 ι§ ’產生兩不重疊的時脈;以及二分別受前述兩時脈 ,一開關和第二開關,其中該第-開關位於第—米二 该電谷與該分享電容間’該第關位 電UT (2) (3) (4) 7 200845546 However, from the above derivation, the damping coefficient ζ will change due to the difference between the output capacitance value and its parasitic resistance value resr. When the output capacitance value c〇 and its parasitic resistance value Resr are very In order to get a sufficient damping coefficient ζ and use a smaller Miller compensation capacitance value Cm2, the second-stage transduction gm2 must be reduced, so that the system feedback gain will be relatively small and the accuracy of the system will be reduced. A trade-off between the damping coefficient control ζ and the system loop gain is formed. SUMMARY OF THE INVENTION • The main purpose of the present invention is to use nested Miller compensation as a basic architecture, and to use the characteristics of pole-column to change the main pole to the output point of the first-stage amplifier. The use of size output capacitors, even zero output capacitors, can also make the system have a good stability advantage, k out in the feedback path of Miller Valley to add active resistors, thereby increasing the damping coefficient control, in response to large The use of the range output capacitor and its parasitic resistance further resolves the trade-off between the damping coefficient control and the system loop gain. Another object of the present invention is to provide a capacitor switching technique to reduce the Miller capacitance required for the entire system, thereby reducing the Miller capacitance required for the entire system, and further extending the frequency because the total capacitance is reduced. Wide and accelerate the settling time of the output voltage. The invention relates to a low-dropout linear voltage regulator, which comprises an input terminal receiving an input DC voltage, and a round output terminal outputting a stable output voltage; a power transistor having a source connected to the wheel-in terminal, and The pole is connected to the output terminal; a first stage amplifier whose inverting input terminal is input with a reference voltage signal by a reference voltage generator, and the non-inverting input terminal is connected to a section 8 200845546 point 'its wheel end and the power transistor a first Miller compensation capacitor is disposed between the drain electrodes; a second stage amplifier having an input terminal connected to the output end of the first stage amplifier, and an output terminal connected to the drain of the power transistor a two-miller compensation capacitor and an active resistor; and a feedback network connected between the drain of the power transistor and the non-inverting input of the first-stage amplifier. The feedback network is composed of two resistors. A voltage divider, and the two resistors form the node. The active resistor is connected to a transistor in the form of a diode. The present invention uses nested Miller compensation as a basic architecture, and utilizes the characteristics of pole-column to change the main pole to the output point of the first-stage amplifier. This method can eliminate the need for large-size output capacitors, and even zero-output capacitors can be used. This makes the system very stable, which is beneficial to the application of the system single chip, and reduces the use of external components and reduces the area of the board. Since the damping coefficient changes due to the difference between the output capacitor value and its parasitic resistance value, the damping coefficient is insufficient, which causes the frequency response to have a surge near the single gain frequency, which in turn causes the output voltage to respond to the load current step 2 The quasi-linear region will cause chopping to occur and slow down its % time. Therefore, the present invention adds f dynamic resistance to the feedback path of the Miller capacitor, thereby increasing the damping coefficient control to cope with the use of a wide range of output electric valleys and parasitic resistance thereof, and further solving the mutual relationship between the damping coefficient control and the system loop gain. The trade-off. The low-dropout linear voltage regulator of the present invention further includes a sharing circuit: the distributed capacitance circuit includes a sharing capacitor, the sharing capacitor privately detects the current on the power transistor, and switches the sharing capacitor respectively The first Miller compensation capacitor or the second Miller compensation capacitor is connected in parallel, thereby reducing the Miller capacitance value required for the entire system, and further accelerating the settling time of the output voltage of 200845546. The '鸪 sharing capacitor circuit includes a current sensing circuit for detecting current on the power transistor; a Schmitt trigger circuit that inputs the signal of the current sensing circuit and transmits the signal to the first The non-overlapping clock generates ι§ 'generating two non-overlapping clocks; and two are respectively subjected to the aforementioned two clocks, a switch and a second switch, wherein the first switch is located between the electric valley and the shared capacitor 'The first level of electricity

容與該分享電HCapacity and the sharing of electricity H

至盥:電晶體操作在三極管區時’該分享電容切換 來Ϊ主極雜電容並聯’行成較Α的米勒補償電容值 在三極管區,該分㈣:亡 力率電晶體操作 並聯,行成較與第—米勒補償電容^ 使系,=:電容值來將主極點推向低頻, 盥第當曰曰體操作在飽和區時,該分享電容切換至 增加阻尼係數的控制 大料勒補彳貝電奋值來 電晶體操作在飽和區,該二持續加’直到該功率 電容並聯’行成較大的米與第二米勒補償 控制,由於此時有較小的二貝電各值來增加阻尼係數的 授路徑,因此當重負載±相:值於第—個米勒補償電容回 【實施方式】負⑽頻寬將得到延展。 成有關本發明之詳細# 說明之用,而 來作進-步說明,但應:t技術說明 ,現以實施例 不應被解釋為本發“ v 解的疋,該等實施例僅為例示 200845546 請參閱「第3圖」所示,係本發明在米勒電容的回授 路徑上加入主動電組之示意圖。使用巢狀米勒補償為基本 架構該低壓降線性電壓穩壓器1〇〇,該低壓降線性電壓穩 壓為100包括一輸入端vIN接收輸入的直流電壓,並有一輸 出端νουτ輸出一穩定的輸出電壓。一功率電晶體13〇及串 聯的一第一級放大器110與一第二級放大器12〇 ;功率電晶 體130的源極與輸入端Vin相連,其汲極與輸出端ν〇υτ相連 接,其閘極與第二級放大器120的輸出端相連;第一級放 _ 大器110的反相輸入端由一參考電壓產生器140輸入一參 考電壓訊號’其同相輸入端連接至一節點15Ό,且其輸出 ^連接至該第二級放大器12〇的輸入端,且輸出端與功率 電晶體130的汲極間的回授路徑上設有一第一米勒補償電 谷Cmi,而且弟一級放大器120的輸出端與功率電晶體;[30 的汲極間的回授路徑上設有一第二米勒補償電容(:^2。一 回授網路200係連接在功率電晶體130的汲極與第一級放 大器11 〇的同相輸入端之間,該回授網路200係包括電阻Rpi 和R η,該兩電阻rfi、r打構成了一分壓器,且兩電阻、 R F2間構成該節點150,第一級放大器110的同相輸入端連 接至該節點150。該低壓降線性電壓穩壓器1〇〇的輸出端 vOUT並外接有一輸出電容Cl與其寄生電阻Resr。 其特徵在於本發明在第二級放大器12〇的輸出端與功 率電晶體130的 >及極間的第二米勒補償電容cm2回授路徑 上设有一主動電阻160與該第二米勒補償電容串聯,藉 此增加阻尼係數的控制以因應大範圍輪出電容cL與其寄 生電阻Resr的使用,並可以進一步解決阻尼係數ζ控制與系 統迴路增益間相互的取捨。其中該主動電阻16〇係連接成 200845546 二極體形式的電晶體。 、因為,由公式(1) (2) (3) (4)的推導可知阻尼係數ζ會因 為輸出電容值cOUT以及其寄生電阻值1^訊的不同而改變, 當輸出電容值cOUT以及其寄生電阻值Resr#常小時,為了 得到一個足夠的阻尼係數ζ和使用較小的米勒補償電容值 Cm2,第一級的轉導gm2必須縮小,如此系統回授增益將會 相對變小而降低了系統的精準度,形成阻尼係數控制 ^ 統迴路增益間相互的取捨。 vTo the 盥: When the transistor is operated in the triode area, 'the shared capacitance is switched to Ϊ the main-pole capacitance is connected in parallel'. The damper Miller compensation capacitance value is in the triode area. The sub-fourth: the dead force rate transistor operation is connected in parallel. The comparison with the first-Miller compensation capacitor ^ makes the system, =: the capacitance value to push the main pole to the low frequency, and when the first body is operated in the saturation region, the shared capacitance is switched to the control of increasing the damping coefficient. In addition to the 彳 电 电 奋 来电 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体 晶体To increase the damping coefficient of the routing path, so when the heavy load ± phase: the value of the first Miller compensation capacitor back [embodiment] negative (10) bandwidth will be extended. In the detailed description of the present invention, the description will be made, but the description should be made with the following description, but the following description should not be construed as an example of the present invention. The embodiments are merely illustrative. 200845546 Please refer to "Figure 3" for a schematic diagram of the present invention for adding an active power pack to the feedback path of a Miller capacitor. Using a nested Miller compensation as the basic architecture of the low-dropout linear voltage regulator 1〇〇, the low-dropout linear voltage regulator is 100 including an input terminal vIN receiving input DC voltage, and an output terminal νουτ output is stable The output voltage. a power transistor 13A and a first-stage amplifier 110 connected in series with a second-stage amplifier 12A; the source of the power transistor 130 is connected to the input terminal Vin, and the drain thereof is connected to the output terminal ν〇υτ, The gate is connected to the output of the second stage amplifier 120; the inverting input of the first stage amplifier 110 is input by a reference voltage generator 140 to a reference voltage signal 'the non-inverting input is connected to a node 15A, and The output is connected to the input end of the second stage amplifier 12〇, and a first Miller compensation electric valley Cmi is disposed on the feedback path between the output end and the drain of the power transistor 130, and the first stage amplifier 120 is The output end is connected to the power transistor; a second Miller compensation capacitor is provided on the feedback path between the drains of the 30 (: ^2. A feedback network 200 is connected to the drain of the power transistor 130 and the first Between the non-inverting input terminals of the stage amplifier 11 ,, the feedback network 200 includes resistors Rpi and R η , and the two resistors rfi and r form a voltage divider, and the two resistors and R F2 form the node 150 . The non-inverting input of the first stage amplifier 110 is connected to the node 15 0. The output terminal vOUT of the low-voltage drop linear voltage regulator 1 并 is externally connected with an output capacitor C1 and its parasitic resistance Resr. The invention is characterized in that the output of the second stage amplifier 12 与 and the power transistor 130 And the second Miller compensation capacitor on the inter-electrode cm2 feedback path is provided with an active resistor 160 in series with the second Miller compensation capacitor, thereby increasing the damping coefficient control in response to a wide range of wheel-out capacitance cL and its parasitic resistance Resr The use of the damping coefficient ζ control and the system loop gain can be further solved. The active resistor 16 is connected to the transistor of the 200845546 diode form, because, by the formula (1) (2) ( 3) The derivation of (4) shows that the damping coefficient ζ will change due to the difference between the output capacitance value cOUT and its parasitic resistance value. When the output capacitance value cOUT and its parasitic resistance value Resr# are often small, in order to get an adequate Damping coefficient ζ and using a smaller Miller compensation capacitor value Cm2, the first stage of the transduction gm2 must be reduced, so the system feedback gain will be relatively small and reduce the accuracy of the system To form a mutual trade-off between damping control system loop gain ^. V

基於以上理由,我們提出在相關的米勒電容的回授路 徑上加入主動電阻160。請再參閱「第4圖」所示的小訊號 模型,該小訊號模型包含了第一級放大器11〇與第二級放 大器120的兩個增盈級gmlvs、gwV2和一個功率電晶體 的輸出級,其中RF1、1^2組成回授網路,gmi、g⑽、§猶、 gmp分別為苐一、第二放大器11〇、12〇、主動電阻和輸 出級的轉導’ R01、1102和CP1、CP2分別為第一級、第二級 放大器110、120的輸出阻抗和寄生電容,c〇uT為輸出電 容,Resr為輸出電容的寄生電阻,cml、Cm2為第一、第二 米勒補償電容,ROUT (=Rl||r〇p|(Rfi+Rfi))為等效輸出 龟阻’其中Rl和R〇p分別為負載阻抗和功率電晶體的輸出 A l + sCm2 V ^ma SmlSmp v ( \ 1+」 、 ^OUT^ESR J ( \ u 5 V P-3dB J ( η Γ λ 1 + 5 CoutResr + — + ~~— +52 _ 、 ^/»2 Sma ) / γ ^ml^OUT ^ SmzSmp J 阻尼係數為ζ = $ (5) 經由推倒解出For the above reasons, we propose to add the active resistor 160 to the feedback path of the associated Miller capacitor. Please refer to the small signal model shown in "Fig. 4". The small signal model includes two gain stages gmlvs, gwV2 of the first stage amplifier 11 and the second stage amplifier 120, and an output stage of a power transistor. , where RF1 and 1^2 form a feedback network, gmi, g(10), § y, gmp are respectively 、1, second amplifier 11〇, 12〇, active resistance and output stage transduction 'R01, 1102 and CP1 CP2 is the output impedance and parasitic capacitance of the first-stage and second-stage amplifiers 110 and 120, respectively, c〇uT is the output capacitance, Resr is the parasitic resistance of the output capacitor, and cml and Cm2 are the first and second Miller compensation capacitors. ROUT (=Rl||r〇p|(Rfi+Rfi)) is the equivalent output turtle resistance' where Rl and R〇p are the load impedance and the output of the power transistor respectively A l + sCm2 V ^ma SmlSmp v ( \ 1+" , ^OUT^ESR J ( \ u 5 V P-3dB J ( η Γ λ 1 + 5 CoutResr + — + ~~— +52 _ , ^/»2 Sma ) / γ ^ml^OUT ^ SmzSmp J damping coefficient is ζ = $ (5) solved by pushing down

SmlSml

SmlSmp (6)SmlSmp (6)

OUT 因此,主動電阻160的轉導gma可以設計的很小來增加 12 200845546 阻尼係數的控制,如此第二級放大器120的轉導gm2不用變 動,所以也不會影響系統回授增益而降低了系統的精準 .度。 請再參閱「第5圖」。再者,為了大大的減少米勒補償 電容值的使用,本發明提出一分享電容切換技術。基於穩 定度的考置’當外接為輕負載(light load )時,尤其需要 一個較大的米勒補償電容值(Cwj來將主極點推向低頻,使 系統有足夠的相角邊際。當外接為重負載(heavy load)時, 尤其需要一個較大的第二米勒補償電容值來增加阻 春 尼係數的控制。因此在輕負載和重負載時,其第一、第二 米勒補償電容有不同的需求,所以本發明進一步 在電路中加入一分享電容電路17Ό,該分享電容電路17〇偵 測該功率電晶體130上的電流,使一分享電容(^„3藉由切換 分別與第一米勒補償電容或第二米勒補償電容 聯,來反應不同負載,如此可以減少整個系統所需的米勒 電容值,而且因為總電容值的減少,所以可以進一步延展 頻寬並加速輸出電壓的穩定時間。 • 該分享電容切換技術的實現係利用一電流感測電路 (Current sensing) 171偵測該功率電晶體130上的電流,並 將其結果驅動一史密特觸發(Schmitt Trigger)電路172和 一非重疊時脈產生器(non-〇verlapping clocks generator) 173。因為史密特觸發電路172擁有遲滯空間(hysteresis) 可以避免切換中的雜訊,並且可以加速暫態響應;非重疊 時脈產生器電路173可以避免產生的兩個時脈%和少2重 疊’使位於第一米勒補償電容Cw7與該分享電容Cw3、第二 米勒補償電容C w 2與該分享電容C w 3間的第一開關S W1和第 13 200845546 一開2 SW2不會同時打開或關閉。 當輕負载時’該功率電晶體13〇操作在三極管 位為高電位且第一開關SW1打開(時心 第一米勒補償至與 來將J極點推向低頻,使系統有足夠的相角邊值 虽負載電流持續加,直到該功率電晶體13()操作在飽Therefore, the transconductance gma of the active resistor 160 can be designed to be small to increase the control of the damping coefficient of 12 200845546, so that the transconductance gm2 of the second-stage amplifier 120 does not change, so the system feedback gain is not affected and the system is lowered. Accuracy. Degree. Please refer to "Figure 5" again. Furthermore, in order to greatly reduce the use of the Miller compensation capacitance value, the present invention proposes a shared capacitance switching technique. Based on the stability test, when the external connection is light load, a larger Miller compensation capacitor value (Cwj is needed to push the main pole to the low frequency, so that the system has enough phase angle margin. When external For heavy load, a larger second Miller compensation capacitor value is needed to increase the control of the resistance factor. Therefore, the first and second Miller compensation capacitors are available for light and heavy loads. Different requirements, so the present invention further adds a shared capacitor circuit 17Ό in the circuit, the shared capacitor circuit 17 detects the current on the power transistor 130, so that a shared capacitor (^3 is switched by the first and the first The Miller compensation capacitor or the second Miller compensation capacitor is coupled to react to different loads, which reduces the Miller capacitance required for the entire system, and because of the reduction in total capacitance, it can further extend the bandwidth and accelerate the output voltage. Stabilization time. • The sharing capacitor switching technique is implemented by using a current sensing circuit 171 to detect the current on the power transistor 130 and The result drives a Schmitt Trigger circuit 172 and a non-〇verlapping clocks generator 173. Since the Schmitt trigger circuit 172 has a hysteresis (hysteresis), it avoids mismatch in the switching. And can accelerate the transient response; the non-overlapping clock generator circuit 173 can avoid generating two clocks % and less 2 overlaps to make the first Miller compensation capacitor Cw7 and the sharing capacitor Cw3, the second Miller The first switch S W1 and the 13 200845546 one open 2 SW2 between the compensation capacitor C w 2 and the sharing capacitor C w 3 are not turned on or off at the same time. When the load is light, the power transistor 13 〇 operates in the triode position. High potential and the first switch SW1 is turned on (the first center Miller is compensated to and the J pole is pushed to the low frequency, so that the system has sufficient phase angle value, although the load current continues to increase until the power transistor 13 () operates In full

^ ^ (Z7Tregion} ^ ^ ^ ^ ^ ^ ^ ^ 1 s wi r ^脈2為鬲電位且第二開關SW2打開),分享電容 插^^至與f二米勒補償電容^並聯,行餘大的米勒 ί貝電Ϊ值來增加阻尼係數的控制,由於此時有較小的電 丨—個米勒補償電容回授路徑,因此當重負載時頻 見將得到延展。 本發明藉由將該分享電容(::^並聯至第一米勒補償電 iCw/或第二米勒補償電容Cw來反應不同負载的穩定度考 L’、i此可以減少整個系統所需的米勒電容值,而且因為 、心,減少,所以可以進一步延展頻寬並加速輸出電壓 的% = %間,此方法經由測試後發現可以減少整個系統所 需的=勒電容值約40%,並且不會影響其穩定度。 明同時參閱「第6-1至6-3圖」與「第7-1至7-3 圖」分別為未啟動分享電容切換技術與啟動分享電容切 換技術的測試結果,並且使用相同的測試環境 為1微法技和心以為1歐姆、(2)(:〇『為i微法拉和及^? 為0·3、1人姆、(3)coc/r為零。未啟動分享電容切換技術 的測試結果分別為「第6>β1圖」穩定時間為8微秒、「第 6_2圖」穩定時間為7微秒、「第心3圖」穩定時間為 200845546 ίο微秒。啟動分享電容切換技術的測試結果分別為「第 7-1圖」穩定時間為6微秒、「第7-2圖」穩定時間為5 微秒、「第7-3圖」穩定時間為5微秒。由以上測試結 果可知,提出的分享電容切換技術除了減少了整個系統 所需的米勒電容值,並可以進一步加速輸出電壓的穩定 時間而且不會影響其穩定度。 惟以上所述者,僅為本發明之較佳實施例而已,當不 能以此限定本發明實施之範圍,即大凡依本發明申請專利 範圍及發明說明内容所作之簡單的等效變化與修飾,皆仍 屬本發明專利涵蓋之範圍内。^ ^ (Z7Tregion} ^ ^ ^ ^ ^ ^ ^ ^ 1 s wi r ^ pulse 2 is the zeta potential and the second switch SW2 is turned on), the shared capacitor is inserted into the parallel with the f-miller compensation capacitor ^, the line is large The Miller 贝 Ϊ Ϊ 来 来 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加 增加The present invention can reduce the stability of different loads by using the shared capacitance (:: parallel to the first Miller compensation electric iCw / or the second Miller compensation capacitance Cw) to reduce the stability of the different loads. The Miller capacitance value, and because of the heart, is reduced, so it can further extend the bandwidth and accelerate the %=% of the output voltage. This method has been found to reduce the required capacitance of the whole system by about 40% after testing. It will not affect its stability. See also "Figures 6-1 to 6-3" and "Figures 7-1 to 7-3" for the test results of the shared capacitor switching technology and the startup shared capacitor switching technology. And use the same test environment for 1 microfarad and heart to think 1 ohm, (2) (: 〇 "is i microfarad and ^? is 0. 3, 1 person, (3) coc / r is zero The test results of the shared capacitor switching technique are not the "6th>β1 map" stabilization time is 8 microseconds, the "6th diagram" stabilization time is 7 microseconds, and the "heart 3 diagram" stabilization time is 200845546 ίο micro Second. The test results of the start sharing capacitor switching technology are respectively "第7-1图" The time is 6 microseconds, the stabilization time of "7-2" is 5 microseconds, and the stability time of "7-3" is 5 microseconds. From the above test results, the proposed shared capacitance switching technique is reduced in addition to the whole. The Miller capacitance value required by the system can further accelerate the settling time of the output voltage without affecting its stability. However, the above is only the preferred embodiment of the present invention, and the present invention cannot be limited thereto. The scope of the invention, that is, the simple equivalent changes and modifications made by the present invention in the scope of the invention and the scope of the invention are still within the scope of the invention.

15 200845546 _ 【圖式簡單說明】 第1圖,係使用巢式米勒補償的低壓降線性電壓穩壓器之 示意圖。 第2圖,係第1圖的小訊號模型。 第3圖,係本發明在米勒電容的回授路徑上加入主動電組 之不意圖。 第4圖,係第3圖的小訊號模型。 第5圖,係本發明在第3圖電路上進一步加入分享電容切 換之不意圖。 9 第6-1至6-3圖,係第3圖未具分享電容切換技術的測試 結果。 第7-1至7-3圖,係第5圖具啟動分享電容切換技術的 測試結果。 【主要元件符號說明】 10、 100 :低壓降線性電壓穩壓器 11、 110 :第一級放大器 12、 120 :第二級放大器 • 13、130 :功率電晶體 14、 140 :參考電壓產生器 15、 150 ··節點 20、200 :回授網路 160 :主動電阻 170 :分享電容電路 171 :電流感測電路 172 :史密特觸發電路 173 :非重疊時脈產生器 16 200845546 cml :第一米勒補償電容 cm2:第二米勒補償電容 :分享電容 SW1 :第一開關 SW2 :第二開關15 200845546 _ [Simple description of the diagram] Figure 1 is a schematic diagram of a low-dropout linear voltage regulator using nested Miller compensation. Figure 2 is a small signal model of Figure 1. Fig. 3 is a schematic view of the present invention in which an active power pack is added to a feedback path of a Miller capacitor. Figure 4 is a small signal model of Figure 3. Fig. 5 is a schematic view of the present invention further incorporating a shared capacitor switching on the circuit of Fig. 3. 9 Figures 6-1 to 6-3, Figure 3 shows the results of the shared capacitor switching technique. Figures 7-1 through 7-3, Figure 5 shows the test results for starting the shared capacitor switching technique. [Main component symbol description] 10, 100: Low dropout linear voltage regulator 11, 110: First stage amplifier 12, 120: Second stage amplifier • 13, 130: Power transistor 14, 140: Reference voltage generator 15 150··Node 20, 200: feedback network 160: active resistor 170: shared capacitor circuit 171: current sensing circuit 172: Schmitt trigger circuit 173: non-overlapping clock generator 16 200845546 cml: first meter Le compensation capacitor cm2: second Miller compensation capacitor: sharing capacitor SW1: first switch SW2: second switch

1717

Claims (1)

200845546 十、申請專利範圍: 1. 一種低壓降線性電壓穩壓器,其包括: 一輸入端接收輸入的直流電壓,及一輸出端輸出一穩定 的輸出電壓; 一功率電晶體,其源極與輸入端相連,其汲極與輸出端 相連接, 一第一級放大器,其反相輸入端由一參考電壓產生器輸 入一參考電壓訊號,其同相輸入端連接至一節點,其輸 出端與該功率電晶體的汲極間設有一第一米勒補償電 容; 一第二級放大器,其輸入端連接該第一級放大器的輸出 端,其輸出端與該功率電晶體的汲極間設有串聯的一第 二米勒補償電容與一主動電阻;以及 一回授網路係連接在該功率電晶體的汲極與第一級放大 器的同相輸入端之間,該回授網路係由兩電阻構成的一 分壓器’且兩電阻間構成該節點。 2. 如申請專利範圍第1項所述之低壓降線性電壓穩壓器,其 中’該主動電阻係連接成二極體形式的電晶體。 3. 如申請專利範圍第1項所述之低壓降線性電壓穩壓器,其 中,該低壓降線性電壓穩壓器進一步包括一分享電容電 路,該分享電容電路包含一分享電容,該分享電容電路 偵測該功率電晶體上的電流,且切換該分享電容分別與 該第一米勒補償電容與第二米勒補償電容其中之一產 生並聯。 4. 如申請專利範圍第3項所述之低壓降線性電壓穩壓器,其 中,該分享電容電路包括: 18 200845546 一電流感測電路,其用以偵測該功率電晶體上的電流; 一史密特觸發電路,其接入該電流感測電路的訊號,並 將訊號傳至一非重疊時脈產生器,產生兩不重疊的時 脈;以及 二分別受前述兩時脈控制的第一開關和第二開關,其中 該第一開關位於第一米勒補償電容與該分享電容間,該 第二開關位於第二米勒補償電容與該分享電容間。 5. 如申請專利範圍第3項所述之低壓降線性電壓穩壓器,其 中,當該功率電晶體操作在三極管區時,該分享電容切 換至與第一米勒補償電容並聯,行成較大的米勒補償電 容值來將主極點推向低頻。 6. 如申請專利範圍第3項所述之低壓降線性電壓穩壓 器,其中,當該功率電晶體操作在飽和區時,該分享 電容切換至與第二米勒補償電容並聯,行成較大的米 勒補償電容值來增加阻尼係數的控制。200845546 X. Patent application scope: 1. A low-dropout linear voltage regulator, comprising: an input terminal receiving an input DC voltage, and an output terminal outputting a stable output voltage; a power transistor, a source thereof The input end is connected, the drain is connected to the output end, a first stage amplifier, the inverting input end is input with a reference voltage signal by a reference voltage generator, and the non-inverting input end is connected to a node, and the output end thereof A first Miller compensation capacitor is disposed between the drains of the power transistor; a second stage amplifier having an input end connected to the output end of the first stage amplifier, and an output terminal connected in series with the drain of the power transistor a second Miller compensation capacitor and an active resistor; and a feedback network connected between the drain of the power transistor and the non-inverting input of the first stage amplifier, the feedback network is composed of two resistors A voltage divider is constructed 'and the two resistors form the node. 2. The low-dropout linear voltage regulator of claim 1, wherein the active resistor is connected to a transistor in the form of a diode. 3. The low dropout linear voltage regulator of claim 1, wherein the low dropout linear voltage regulator further comprises a shared capacitor circuit, the shared capacitor circuit comprising a shared capacitor, the shared capacitor circuit Detecting a current on the power transistor, and switching the shared capacitor to be in parallel with one of the first Miller compensation capacitor and the second Miller compensation capacitor, respectively. 4. The low-dropout linear voltage regulator according to claim 3, wherein the shared capacitor circuit comprises: 18 200845546 a current sensing circuit for detecting current on the power transistor; a Schmitt trigger circuit that connects the signal of the current sensing circuit and transmits the signal to a non-overlapping clock generator to generate two non-overlapping clocks; and the second is controlled by the two clocks respectively And a switch, wherein the first switch is located between the first Miller compensation capacitor and the shared capacitor, and the second switch is located between the second Miller compensation capacitor and the shared capacitor. 5. The low-dropout linear voltage regulator of claim 3, wherein when the power transistor is operated in a triode region, the shared capacitor is switched in parallel with the first Miller compensation capacitor. The large Miller compensates for the capacitance value to push the main pole to the low frequency. 6. The low dropout linear voltage regulator of claim 3, wherein when the power transistor is operated in a saturation region, the shared capacitor is switched in parallel with the second Miller compensation capacitor. The large Miller compensates for the capacitance value to increase the damping factor control. 1919
TW96115423A 2007-05-01 2007-05-01 Low dropout (LDO) linear voltage regulator TW200845546A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW96115423A TW200845546A (en) 2007-05-01 2007-05-01 Low dropout (LDO) linear voltage regulator

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW96115423A TW200845546A (en) 2007-05-01 2007-05-01 Low dropout (LDO) linear voltage regulator

Publications (2)

Publication Number Publication Date
TW200845546A true TW200845546A (en) 2008-11-16
TWI329967B TWI329967B (en) 2010-09-01

Family

ID=44822888

Family Applications (1)

Application Number Title Priority Date Filing Date
TW96115423A TW200845546A (en) 2007-05-01 2007-05-01 Low dropout (LDO) linear voltage regulator

Country Status (1)

Country Link
TW (1) TW200845546A (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI386772B (en) * 2009-10-19 2013-02-21 Anpec Electronics Corp Switching voltage regulator
TWI407288B (en) * 2009-05-14 2013-09-01 Sanyo Electric Co Power source circuit
TWI456425B (en) * 2012-01-03 2014-10-11 Univ Nat Taipei Technology Method and system for generating small signal model of voltage converter, computer program product

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI730534B (en) * 2019-12-09 2021-06-11 大陸商北京集創北方科技股份有限公司 Power supply circuit and digital input buffer, control chip and information processing device using it

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI407288B (en) * 2009-05-14 2013-09-01 Sanyo Electric Co Power source circuit
TWI386772B (en) * 2009-10-19 2013-02-21 Anpec Electronics Corp Switching voltage regulator
TWI456425B (en) * 2012-01-03 2014-10-11 Univ Nat Taipei Technology Method and system for generating small signal model of voltage converter, computer program product

Also Published As

Publication number Publication date
TWI329967B (en) 2010-09-01

Similar Documents

Publication Publication Date Title
US20090001953A1 (en) Low dropout linear voltage regulator
TWI517544B (en) Current sensing apparatus, current sensing system and current digital-to-analog converter
TWI441006B (en) On-chip low voltage capacitor-less low dropout voltage regulator with q-control, method for forming the same and system for the same
TWI332134B (en) Adaptive pole and zero & pole zero cancellation control low drop-out voltage regulator
US20200274447A1 (en) Power converters and compensation circuits thereof
JP4236586B2 (en) Low dropout voltage regulator
CN101471632B (en) Self-bias low-voltage operation transconductance amplifier circuit with controllable loop gain
TW201618454A (en) Multi-stage amplifier
TWI526801B (en) Improved compensation circuit and the application of its switching power supply
CN105896969A (en) A system and circuit for achieving bidirectional hysteretic current mode control with constant hysteresis
JP2011113567A (en) Low-dropout linear regulator (ldo), method for providing ldo and method for operating ldo
JP2012125107A (en) Control circuit for switching regulator, switching regulator using the same, and electronic apparatus
CN103838287A (en) Linear voltage regulator for compensation zero point dynamic adjustment
JP5232306B2 (en) Control circuit for DC-DC converter
CN103490616A (en) Current-mode DC-DC converter
JP2017512341A (en) Low dropout voltage regulator circuit
TW200845546A (en) Low dropout (LDO) linear voltage regulator
CN107305399B (en) PMOS power electric crystal linear voltage decreasing regulator circuits
JP2002032133A (en) Regulated power supply circuit
TWI400592B (en) Low dropout regulator
CN113625810A (en) Low-power-consumption full-range stable LDO linear voltage regulator without off-chip capacitor
CN205353832U (en) Need not outer capacitance compensation's of piece high transient response linear voltage regulator
CN106155155A (en) One single-transistor low dropout voltage regulator
Wang et al. A high stability low drop-out regulator with fast transient response
CN221305757U (en) Structure for improving loop response based on III-type compensation network

Legal Events

Date Code Title Description
MM4A Annulment or lapse of patent due to non-payment of fees