The present application claims priority to India Provisional Patent Application No. 202241034499, which was filed Jun. 16, 2022, is titled “PSEUDO ESR TECHNIQUE FOR MULTI-LOOP FLIPPED-VOLTAGE-FOLLOWER LDO WITH REVERSE CURRENT PROTECTION,” and is hereby incorporated herein by reference in its entirety.
Linear voltage regulators are useful in power systems to receive a variable input voltage and to provide a stable, low-noise power supply. A low dropout (LDO) linear voltage regulator is a type of linear voltage regulator circuit that works well even when the output voltage is close to the input voltage, resulting in improved power efficiency over conventional regulators. A feedback loop may be useful to stabilize the LDO. If the supply and output voltages are higher than five to seven volts, the feedback loop may include a high voltage capacitor for stabilizing the LDO.
In accordance with at least one example of the disclosure, a voltage regulator includes a first feedback loop. The first feedback loop includes a first transistor coupled to an input voltage terminal and a second transistor coupled to the first transistor. The first feedback loop also includes a third transistor coupled to the second transistor, and a fourth transistor coupled to the third transistor and a gate of the first transistor. The voltage regulator includes a resistor having a first terminal and a second terminal, the first terminal coupled to the first transistor and the second transistor. The voltage regulator also includes a second feedback loop, where the second feedback loop includes an amplifier having a first amplifier input coupled to the second terminal of the resistor, a second amplifier input coupled to a voltage source, and an amplifier output coupled to a gate of the second transistor.
In accordance with at least one example of the disclosure, a voltage regulator includes a first feedback loop, where the first feedback loop includes a pass transistor coupled to an input voltage terminal and a sense transistor coupled to the pass transistor. The first feedback loop also includes a control transistor coupled to the sense transistor and a cascode transistor coupled to the control transistor. The first feedback loop includes a buffer transistor coupled to the cascode transistor and a gate of the sense transistor. The voltage regulator includes a resistor having a first terminal and a second terminal, the first terminal coupled to the sense transistor and the control transistor. The voltage regulator also includes a second feedback loop, where the second feedback loop includes an amplifier having a first amplifier input coupled to the second terminal of the resistor, a second amplifier input coupled to a voltage source, and an amplifier output coupled to a gate of the control transistor.
In accordance with at least one example of the disclosure, a low-dropout voltage regulator includes a pass transistor having a flipped-voltage-follower configuration with a control transistor, where a drain of the pass transistor is configured to provide an output voltage of the low-dropout voltage regulator. The low-dropout voltage regulator includes a first feedback loop for the voltage regulator, where the first feedback loop is configured to provide a load transient response. The first feedback loop includes the pass transistor and the control transistor. The first feedback loop also includes a cascode transistor coupled to the control transistor. The first feedback loop includes a buffer transistor coupled to the cascode transistor and a gate of the pass transistor. The low-dropout voltage regulator includes a resistor having a first terminal and a second terminal, the first terminal coupled to the pass transistor and the control transistor. The low-dropout voltage regulator also includes a second feedback loop for the voltage regulator, where the second feedback loop is configured to provide load regulation. The second feedback loop includes an amplifier having a first amplifier input coupled to the second terminal of the resistor, a second amplifier input coupled to a voltage source, and an amplifier output coupled to a gate of the control transistor.
The same reference numbers or other reference designators are used in the drawings to designate the same or similar (functionally and/or structurally) features.
An LDO may include one or more feedback loops that stabilize the LDO for a wide range of effective series resistance (ESR). The ESR is the resistance of a non-ideal capacitor, and the ESR occurs in series with the capacitor. The ESR may cause a variety of performance issues in an electronic circuit, such as power losses, noise, voltage drops, etc. To compensate for the ESR, one LDO includes an additional resistor coupled to a pass transistor in the load current path and samples feedback both before and after the resistor. A transfer function describes the LDO operation. The transfer function is a function that models the system's output for each input (e.g., VOUT/VIN). A transfer function may have one or more zeros, which is the frequency value for which the value of the numerator of the transfer function becomes zero. A transfer function may also have one or more poles, which is the frequency value for which the value of the denominator of the transfer function becomes zero. The additional resistor in this LDO adds a zero to the transfer function of the circuit. This zero tracks the pole occurring due to the output capacitance, and the additional resistor bounds the upper frequency limit of the zero to help stabilize the circuit. However, the LDO uses a high voltage capacitor in at least one of the feedback paths, which increases the area of the circuit.
In another LDO, the additional resistor is removed from the load current path, and a series resistance is realized using a sense transistor coupled to the pass transistor of the LDO. A capacitor coupled to the sense resistor provides the feedback in a feedback loop. Compared to the above described LDO, this LDO removes the resistor in series with the pass transistor. This LDO also includes the high voltage capacitor in the feedback path, which increases the area of the circuit.
In examples herein, the capacitor in the feedback path of the LDO is removed and replaced with an amplifier. To achieve a low quiescent current (IQ) and good transient performance in an LDO, a flipped-voltage-follower (FVF) feedback loop is added along with a main feedback loop having a high gain error amplifier to maintain high accuracy at the output. An FVF is a buffer cell with a local feedback loop. In an FVF, a voltage output is provided at a drain of a pass transistor, which is coupled to a source of a control transistor. The drain of the control transistor is coupled to the gate of the pass transistor. A current source is also coupled to the drain of the control transistor. The FVF has low impedance at the voltage output, and can source currents much larger than the current source.
The two feedback loops described above allow feedback to be taken both before and after an additional resistor coupled to the pass transistor. The additional resistor is referred to as an intentional resistor (RINT) or an FVF resistor (RFVF). The two feedback loops include a high speed loop and a low speed loop. The high speed loop (e.g., the FVF loop) includes a high bandwidth amplifier for fast load transient response. The high speed loop decreases the effective output impedance due to negative feedback gain. Therefore, with higher load currents, the output pole moves to a higher frequency. The low speed loop includes a high direct current (DC) gain amplifier for good load regulation. The low-speed loop maintains output accuracy by sensing the feedback at the output node of the LDO. The RINT adds a left-half plane (LHP) zero in the loop transfer function of the high speed loop that scales with the output capacitance, which helps stabilize the high speed loop by bounding the frequency movement of the zero created by the ESR. The RINT acts like an ESR for the high speed loop without adding a zero to the low speed loop. The RINT limits the lowest impedance on the output of the LDO, which sets an upper bound on the output pole frequency and therefore prevents complex pole pair formation at high load currents. The structure described herein provides good load regulation and better stability without a high voltage capacitor in a feedback loop that would increase area.
In another example, a sense transistor is added to avoid the use of a resistor (e.g., RINT or RFVF) in the load current path. The sense transistor injects a current proportional to the load through a resistor, which introduces a zero in the loop that tracks the output pole. In another example, to boost the phase margin of the LDO at full loads, the ESR zero should be moved to higher frequencies as the load increases. To move the ESR zero to higher frequencies with a load, the pass transistor to sense transistor ratio should increase with increases in the load. This increase may be realized by adding a degeneration resistance coupled between the sense transistor and a voltage supply.
First amplifier 102 includes a first input 118 that receives a reference voltage VREF, and a second input 120 that is coupled to output node 114 (VOUT). First amplifier 102 includes an amplifier output 122. Amplifier output 122 is coupled to first input 124 of second amplifier 104. Second amplifier 104 has a second input 126 that is coupled to RINT 106. Second input 126 is also coupled to amplifier output 128 of second amplifier 104. LDO 100 includes a first feedback loop 130 and a second feedback loop 132. First feedback loop 130 is the low speed loop as described above. First feedback loop 130 maintains good load regulation, so feedback is taken from the output at node 114. Second feedback loop 132 provides a fast load transient response. The use of second amplifier 104 in second feedback loop 132 replaces a capacitor in the feedback loop from other systems.
LDO 200 includes a pass transistor 202 (MP), control transistor 204 (MC), buffer transistor 206 (MB), cascode transistor 208 (MCAS), and reverse current protection (RCP) transistor 210 (MRCP). LDO 200 also includes current source 212 (IFVF) and current source 214 (IB). LDO 200 also includes capacitor 216 (C1), resistor 218 (R1), and resistor 220 (R2). Capacitor 216, resistor 218, and resistor 220 may comprise an active load in one example. An active load may have a resistance that changes as overall load current increases. LDO 200 also includes a transconductance amplifier 222 and a set of feedback capacitors and resistors: resistors 224 (RF1), 226 (RF2), 228 (RF3), and capacitors 230 (CF1), 232 (CF2), and 234 (CF3). Transconductance amplifier 222 includes a first input 236, a second input 238, and an output 240. First input 236 receives a reference signal (ADJ_REF) and second input is coupled to output node 114 as part of first feedback loop 130. ADJ_REF is a voltage reference that defines the output voltage value VOUT. ADJ_REF may be provided by an external voltage source or from an internal bandgap circuit (not shown in
LDO 200 includes several nodes, such as nodes 242, 244, 246, 248, 250, 252, and 254. LDO 200 also includes a voltage terminal 256 (e.g., input voltage terminal) that provides a voltage VIN.
First feedback loop 130 and second feedback loop 132 are shown in LDO 200. First feedback loop 130 is the low speed loop, also referred to as the error amplifier loop. First feedback loop 130 connects output node 114 (VOUT) to transconductance amplifier 222. First feedback loop also includes a high impedance node 252 coupled to output 240, and then continues to the gate of control transistor 204. Control transistor 204 acts as a source follower for the first feedback loop 130. The gate to source of control transistor 204 has unity gain. The gain for first feedback loop 130 comes from the transconductance amplifier 222. Feedback for the first feedback loop 130 is taken after the RINT 106, which is also referred to here as RFVF 106 (flipped voltage follower). The RFVF 106 separates the feedback points for the feedback loops 130 and 132.
Second feedback loop 132 is the high speed feedback loop. Second feedback loop 132 has an FVF configuration, and acts as a buffer. The FVF configuration boosts transient response speed. In this example, the second feedback loop 132 has unity gain. Second feedback loop 132 begins at node 242, goes through control transistor 204, through cascode transistor 208, through buffer transistor 206, and then through pass transistor 202. Second feedback loop 132 responds during load transients. The high gain stage of first feedback loop 130 is cascaded with the FVF buffer of second feedback loop 132 to achieve good load regulation and transient response. First feedback loop 130 and second feedback loop 132 are independently stabilized for all load conditions. The RFVF 106 acts as an ESR for second feedback loop 132, while not adding a zero to the first feedback loop 130. Additional discussion of the second feedback loop 132 is provided below.
LDO 200 receives an input voltage VIN and provides an output voltage VOUT at output node 114. Pass transistor 202 provides a voltage at node 242. Node 242 is coupled to one terminal of RFVF 106, while the other terminal of RFVF 106 is coupled to output node 114. Current source 212 provides a bias current to bias control transistor 204. Current source 214 provides a bias current to bias buffer transistor 206. A bias voltage VB1 is provided to a gate of reverse current protection transistor 210 by a first bias voltage source, and a bias voltage VB2 is provided to a gate of cascode transistor 208 by a second bias voltage source.
In an example, pass transistor 202 has a gate coupled to a source of buffer transistor 206, a drain coupled to RFVF 106 and to a source of control transistor 204, and a gate coupled to a source of buffer transistor 206. Buffer transistor 206 acts as a buffer that drives pass transistor 202. As described below, pass transistor 202 may create a pole in the transfer function. Buffer transistor 206 pushes the pole of pass transistor 202 to a higher bandwidth, which increases the bandwidth of the LDO.
Control transistor 204 has a source coupled to RFVF 106, a drain coupled to a source of cascode transistor 208 and to current source 212, and a gate coupled to output 240 of transconductance amplifier 222. Cascode transistor 208 has a gate coupled to a voltage source that produces a bias voltage VB2, and a drain coupled to node 250 and to a gate of buffer transistor 206.
Buffer transistor 206 includes a gate coupled to the drain of cascode transistor 208, a drain coupled to the source of pass transistor 202, and a source coupled to the gate of pass transistor 202. The source of buffer transistor 206 is also coupled to current source 214. Reverse current protection transistor 210 has a source coupled to voltage terminal 256 that provides a voltage VIM a gate coupled to a voltage source that provides a bias voltage VB1, and a drain coupled to the source of pass transistor 202.
In operation, current source 212 biases control transistor 204, and VB1 biases reverse current protection transistor 210. Current travels from voltage terminal 256 (e.g., VB1) through reverse current protection transistor 210, pass transistor 202, and then to RLOAD 112. The FVF structure of pass transistor 202 and control transistor 204 provides an output voltage at node 242, which is a low impedance node. LDO 200 can provide an output voltage VOUT that is close to the input voltage VIN. First feedback loop 130 provides feedback from output node 114 (VOUT) to an error amplifier (transconductance amplifier 222). Transconductance amplifier 222 has a high impedance node 252 at its output 240, which is coupled to the gate of control transistor 204. The feedback from first feedback loop 130 is provided to the gate of control transistor 204, which in turn affects the current through pass transistor 202 and control transistor 204. This feedback provides stability to RLOAD 112.
As described above, the transfer function of the LDO 200 may have one or more poles and zeros. The frequency location of the poles and zeros is managed via the circuit design to stabilize the circuit. The poles and zeros should be designed to provide the circuit with a proper phase margin (e.g., the phase when the circuit reaches unity gain). The poles and zeros should also allow the circuit to have an appropriate bandwidth. The circuit should also be stable during operating conditions and the design should also attempt to avoid the formation of complex poles. The design consideration for managing the poles and zeros of the LDO 200 is described below.
In an example herein, pass transistor 202 may be a first transistor and control transistor 204 may be a second transistor. Cascode transistor 208 may be a third transistor and buffer transistor 206 may be a fourth transistor. Reverse current protection transistor 210 may be a fifth transistor.
Current source 212 (IFVF) provides a bias current to bias control transistor 204. Cascode transistor 208 provides protection for the circuit. Load 302 (ZB) is an active load, which means the resistance component of load 302 could change as overall load current increases. A buffer drives the pass transistor 202, where the buffer is buffer transistor 206. Pass transistor 202 may be a large transistor that produces a pole in the transfer function. Buffer transistor 206 pushes the pole of the pass transistor 202 further out in frequency to increase the stability of the circuit, as described below.
In this example, the first pole and zero combination in the transfer function is provided by the ESR (RESR 110) and load capacitor 108. The second pole and zero combination comes from the active load 302, which has a buffer capacitance 304 (CB), a series resistance 306 (RB3), and the active load component. The active load component RB is a combination of resistance 308 (RB1) and Ractive. Ractive is resistance 310 (RB2) in series with 1/gm of transistor 312 (Mactive), where 1/gm is the resistance of the transistor 312. Therefore, the active load component RB is RB1∥Ractive.
Various poles and zeros are labeled in circuit 400. These are the poles and zeros of the transfer function for circuit 400. A pole 406 (PPOW) is created by pass transistor 202. Load pole 408 (PL) and load zero 410 (PZ) are found at output node 114. Buffer pole 412 (PB) and buffer zero 414 (BZ) are found at the gate of buffer transistor 206. This system has three poles and two zeros.
In curve 502 (without RESR 110), a load pole 408 (PL) occurs, and then the next pole is buffer pole 506 (PB). The load zero 410 (ZL) is not present without the RESR 110, so no zero exists in curve 502 between load pole 408 and buffer pole 506. In curve 504, the RESR 110 adds the load zero 410 (ZL). Therefore, curve 504 represents a pole-zero-pole system. In this pole-zero-pole system, the gain is flat until load pole 408, then the gain begins to drop. At the location of load zero 410, the gain flattens until the location of buffer pole 412. Then, at buffer pole 412, the gain begins to fall again. The pole-zero-pole system with RESR 110 has a phase margin of 90 degrees (e.g., the phase when the system reaches 0 dB or unity gain). Another feature of the system with RESR 110 is that the bandwidth is extended. Curve 504 has greater bandwidth compared to curve 502. An LDO with a fast loop (second feedback loop 132) as described herein avoids complex pole formation. Complex poles are often detrimental to stability and are usually avoided.
Because curve 504 provides these advantages compared to curve 502, it is advantageous for the load zero 410 to occur between load pole 408 and buffer pole 412. Therefore, the circuit may be designed to bound the movement of the load zero 410 to produce a stable system with high bandwidth. The process for bounding the load zero 410 is described below.
In circuit 600, a reverse current protection transistor 210 is coupled in series with the pass transistor 202. As described above with respect to
As discussed above with respect to
If the circuit has an RESR 110 that is nonzero, that ESR is added to the value of RFVF 106, which could slightly shift the location of left half plane zero 602. However, left half plane zero 602 will stay between load pole 408 and buffer pole 412.
This example addresses the problems described above because the circuit described in
As described above, the error amplifier loop (first feedback loop 130) is a high gain and low bandwidth loop that provides excellent tracking accuracy. The dominant pole is the output pole of the error amplifier, and the loop is stabilized with the help of a constant phase compensation network. The pseudo ESR (e.g., RFVF 106) described above stabilizes the second feedback loop 132. In first feedback loop 130, the feedback is taken directly from the output capacitor node (e.g., output node 114 in
The damping factor measures the separation of the complex poles. The damping factor is at a minimum for a low output capacitance, because the load pole 408 is closer to the buffer pole 412 at no load. If the damping factor is less than 1.0, complex poles form. If the damping factor falls below 0.7, the complex poles result in gain peaking, which may result in instability. Curve 1206 shows the example damping factor with no ESR (0 Ohms) in the system. As the load current increases, curve 1206 falls below a 1.0 damping factor, and then continues to fall below 0.5. Curve 1206 would produce gain peaking in the system, which indicates a complex pole.
Curve 1204 corresponds to a system with an ESR of 0.5 Ohms. With an ESR of 0.5 Ohms, the damping factor falls below 1.0 but stays above 0.7, so gain peaking may not occur. Curve 1206 corresponds to a system with an ESR of 3 Ohms. The bandwidth of this loop is high enough that complex poles will not form (e.g., the damping factor stays above 1.0). This example produces a second order system with the poles separated.
Therefore, in some examples, having a pseudo ESR of 0.5 Ohms or more (which is added to the total ESR of the system) provides a higher damping ratio and ensures that there will be no gain peaking in the system. This is because the FVF buffer pushes out the load pole in frequency as the load current increases by reducing the output impedance. Adding a pseudo ESR (e.g., with RFVF 106) ensures a minimum impedance of RFVF looking into the buffer. This impedance limits how far the load pole can be pushed out and stops the formation of complex poles (e.g., PL<PB with RFVF in the circuit).
The first feedback loop 130 output pole (load pole 408) is dominant because high DC gain is needed to achieve good output accuracy. A low frequency constant phase network is used to boost phase while attenuating the gain so that the overall loop bandwidth is limited. As described above, the load pole 408 is pushed out to a high frequency as load current increases. For a small output capacitance, load pole 408 may move out and form a complex pole pair with buffer pole 412. However, the movement of load pole 408 is limited by adding RFVF 106 as a pseudo ESR, which keeps the load pole 408 at a lower frequency than the buffer pole 412 in all cases. The value of RFVF 106 is chosen such that the load pole 408 stops before the buffer pole 412 for a low output capacitance (e.g., 0.5 μF) to ensure no complex pole formation. For a significant load capacitance, an additional zero may exist between load pole 408 and buffer pole 412. The constant phase network ensures that the gain intersects the UGB frequency with −30 dB per decade slope in the worst case scenario. This ensures a minimum phase margin of 45 degrees.
In the examples described herein, an RVFV 106 is added to the LDO, and feedback is taken on each side of RFVF 106. A voltage drop occurs across RFVF 106. There is also a voltage drop from VINT at voltage terminal 256 across the reverse current protection transistor 210, and another voltage drop across pass transistor 202. These voltage drops add up to the total voltage drop between VIN at voltage terminal 256 and VOUT at output node 114.
To avoid the use of a resistor (RFVF 106) in the load current path, sense transistor 1402 is added to inject a current proportional to the load through RFVF 106, which introduces a zero in the loop that tracks the output pole. Sense transistor 1402 mimics the pass transistor 202. Any change in voltage across pass transistor 202 is mirrored with sense transistor 1402. That voltage change is captured through the control transistor 204 and fed back via a feedback loop. The feedback loop in this example includes sense transistor 1402 and the control transistor 204. The effects of RFVF 106 are realized without having RFVF 106 in the path of the pass transistor 202.
The RFVF 106 resistive value realized in circuit 1400 may be found via equation (1):
In equation (1), the RINT value is multiplied by the ratios of the width and lengths of the sense transistor 1402 and the pass transistor 202 to get RFVF 106. WP is the width of pass transistor 202, and LP is the length of pass transistor 202. WS is the width of sense transistor 1402, and Ls is the length of sense transistor 1402.
Circuit 1500 includes the pass transistor 202 and sense transistor 1402 described above with respect to
In examples herein, a sense transistor and/or a resistor may be added to generate a pseudo ESR in a flipped-voltage-follower LDO loop that enables zero to high ESR support. The added sense transistor and resistor structure in the FVF loop avoids complex pole formation in the high gain error amplifier loop, which allows a large resistive value of a buffer zero that enables a low IQ FVF loop design. The right half plane zero introduced due to a reverse current protection transistor is compensated by the added left half plane pseudo ESR zero.
In examples herein, the high-speed FVF loop is stabilized for zero to high ESR. The examples herein avoid complex pole gain peaking without increasing IQ in the FVF loop, which leads to an overall low IQ design. The right half plane zero effect from a reverse current protection transistor may be nullified without degrading transient performance. In one solution, a resistor is not present in the load current path, so the voltage drop across the resistor does not add to dropout voltage. In addition, no high voltage capacitors are used in the implementations, which reduces overall area of the circuit.
In this description, the term “couple” may cover connections, communications, or signal paths that enable a functional relationship consistent with this description. For example, if device A generates a signal to control device B to perform an action: (a) in a first example, device A is coupled to device B by direct connection; or (b) in a second example, device A is coupled to device B through intervening component C if intervening component C does not alter the functional relationship between device A and device B, such that device B is controlled by device A via the control signal generated by device A.
A device that is “configured to” perform a task or function may be configured (e.g., programmed and/or hardwired) at a time of manufacturing by a manufacturer to perform the function and/or may be configurable (or reconfigurable) by a user after manufacturing to perform the function and/or other additional or alternative functions. The configuring may be through firmware and/or software programming of the device, through a construction and/or layout of hardware components and interconnections of the device, or a combination thereof.
A circuit or device that is described herein as including certain components may instead be coupled to those components to form the described circuitry or device. For example, a structure described as including one or more semiconductor elements (such as transistors), one or more passive elements (such as resistors, capacitors, and/or inductors), and/or one or more sources (such as voltage and/or current sources) may instead include only the semiconductor elements within a single physical device (e.g., a semiconductor die and/or integrated circuit (IC) package) and may be coupled to at least some of the passive elements and/or the sources to form the described structure either at a time of manufacture or after a time of manufacture, for example, by an end-user and/or a third-party.
While certain components may be described herein as being of a particular process technology, these components may be exchanged for components of other process technologies. Circuits described herein are reconfigurable to include the replaced components to provide functionality at least partially similar to functionality available prior to the component replacement. Components shown as resistors, unless otherwise stated, are generally representative of any one or more elements coupled in series and/or parallel to provide an amount of impedance represented by the shown resistor. For example, a resistor or capacitor shown and described herein as a single component may instead be multiple resistors or capacitors, respectively, coupled in parallel between the same nodes. For example, a resistor or capacitor shown and described herein as a single component may instead be multiple resistors or capacitors, respectively, coupled in series between the same two nodes as the single resistor or capacitor.
Uses of the phrase “ground voltage potential” in the foregoing description include a chassis ground, an Earth ground, a floating ground, a virtual ground, a digital ground, a common ground, and/or any other form of ground connection applicable to, or suitable for, the teachings of this description. In this description, unless otherwise stated, “about,” “approximately” or “substantially” preceding a parameter means being within +/−10 percent of that parameter. Modifications are possible in the described examples, and other examples are possible within the scope of the claims.
Number | Date | Country | Kind |
---|---|---|---|
202241034499 | Jun 2022 | IN | national |