AMPLIFIER CIRCUIT WITH DYNAMIC OUTPUT SLOPE COMPENSATION FOR DRIVING LARGE CAPACITIVE LOADS

Abstract

An amplifier circuit is described herein. In accordance with one embodiment, the circuit includes an input stage and an output stage. The input stage has a non-inverting input and an inverting input for receiving a differential input voltage and is configured to provide an output signal that represents the differential input voltage. The output stage is configured to receiveas input signalthe output signal of the input stage and to provideat an amplifier outputan output voltage based on the input signal. A feed-back path couples the amplifier output with the inverting input of the input stage. A feed-forward circuit is configured to activate a current path coupled to the amplifier output to provide additional output current when the differential input voltage crosses a threshold.

Claims

1. An amplifier circuit comprising: an input stage including a non-inverting input and an inverting input for receiving a differential input voltage, the input stage being configured to provide an output signal that represents the differential input voltage; an output stage configured to receive, as an input signal, the output signal of the input stage and further configured to provide, at an output, an output voltage based on the input signal; a feed-back path that couples the output of the output stage with the inverting input of the input stage; and a feed-forward circuit is configured to activate a current path coupled to the output to provide an additional output current when the differential input voltage crosses a threshold value.

2. The amplifier circuit of claim 1, wherein the inverting input is connected to a first input node via a first resistor and the non-inverting input is connected to a second input node via a second resistor, and wherein the second input node and the first input node are connected by a current sense resistor.

3. The amplifier circuit of claim 2, wherein the feed-back path includes a third resistor.

4. The amplifier circuit of claim 3, wherein the non-inverting input is connected to a reference voltage source via a fourth resistor.

5. The amplifier circuit of claim 1, wherein the current path of the feed-forward circuit includes a controllable current source coupled to the output of the output stage, wherein the controllable current source is activated when the differential input voltage crosses the threshold value.

6. The amplifier circuit of claim 1, wherein the output stage includes a source follower.

7. The amplifier circuit of claim 6, wherein the source follower comprises an n-channel field effect transistor, and wherein the feed-forward circuit is configured to sink the additional output current when the differential input voltage falls below the threshold value.

8. The amplifier circuit of claim 6, wherein the source follower comprises a p-channel field effect transistor, and wherein the feed-forward circuit is configured to source the additional output current when the differential input voltage exceeds the threshold value.

9. A method comprising: providing, by an input stage of an amplifier, an output signal that represents a differential input voltage of the input stage; providing, by an output stage of the amplifier, an output voltage based on the output signal of the input stage at an amplifier output, wherein a feed-back path couples the amplifier output with an inverting input of the input stage; and activating a current path coupled to the amplifier output to provide an additional output current when the differential input voltage crosses a threshold value.

10. A current sense amplifier comprising: a first input node and a second input node configured to sense a voltage drop across a current sense resistor; an input stage including an inverting input, which is coupled to the first input node via a first resistor, and a non-inverting input, which is coupled to the second input node via a second resistor, wherein the input stage is configured to provide an output signal that represents a differential input voltage received at the inverting input and the non-inverting input; an output stage that includes a source follower configured to receive, as input signal, the output signal of the input stage and further configured to provide, at an amplifier output, an output voltage based on the input signal; a feed-back path that couples the output of the output stage with the inverting input of the input stage; and a feed-forward circuit is configured to activate a current path coupled to the output to provide an additional output current when the differential input voltage crosses a threshold value.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In the following detailed description, reference is made to the accompanying drawings. The drawings form a part of the description and, for the purpose of illustration, show examples of how the invention may be used and implemented. It is to be understood that the features of the various embodiments described herein may be combined with each other, unless specifically noted otherwise.

[0010] FIG. 1 illustrates one example of a current sense amplifier circuit.

[0011] FIG. 2 illustrates an example of an output stage which may be used in the circuit of FIG. 1.

[0012] FIG. 3 illustrates one example of an improved current sense amplifier circuit, which has an improved dynamic performance while requiring only a low quiescent current even for higher output capacitances.

[0013] FIG. 4 illustrates an alternative implementation of the embodiment of FIG. 3.

[0014] FIG. 5 illustrates the complementary circuit with respect to the circuit of FIG. 4.

DETAILED DESCRIPTION

[0015] Analog current sensing solutions that are capable of driving high capacitive loads at low quiescent currents can be implemented using a class AB output stage in the difference amplifier which is used in the current sense amplifier circuit. However, solutions using a class AB output stage may have a reduced linearity, problems concerning the loop stability and/or a higher transient response time. Loop stability or transient response time could be affected because using a class AB output stage usually implies using a Miller compensation.

[0016] Simultaneously satisfying the requirements concerning linearity (distortion), loop stability and transient response time may therefore be a big challenge when using class AB output stages in the difference amplifier. Therefore, some embodiments described herein, use a source follower output stage. However, without further measures (which will be described later), source followers usually allow only small output capacitances. Larger output capacitances such as 2-3 nF would require an undesired (or unfeasible) increase in quiescent current.

[0017] The embodiments described herein may enable a fully integrated solution for a current sense amplifier circuit which is capable to ensure a low static (DC) current consumption while providing a fast response time to a large input differential voltage jump for a wide range of capacitive loads (e.g. 10 pF to 3 nF).

[0018] FIG. 1 illustrates, by way of example, the general structure of a current sense amplifier circuit, which is configured to amplify the voltage drop Vs across the current sense resistor R.sub.S (Vs=i.sub.S.Math.R.sub.S), which carries the current is to be measured. The current sense resistor R.sub.S is connected between a first input node ISN and a second input node ISP of the current sense amplifier circuit.

[0019] The core of the current sense amplifier circuit is a differential amplifier which comprises an input stage AMP.sub.1 and output stage AMP.sub.2. The input stage AMP.sub.1 has a non-inverting input (+) and an inverting input (?). These inputs are configured to receive a differential input voltage V.sub.P?V.sub.N, wherein V.sub.P denotes the voltage at the non-inverting input and V.sub.N denotes the input of the inverting input. The input stage AMP.sub.1 is configured to provide an output signal V.sub.O that represents the differential input voltage V.sub.P?V.sub.N. The input stage AMP1 is a differential amplifier with a high open-loop gain (e.g. greater than 90 dB). In many implementations, the input stage includes an operational transconductance amplifier (OTA). The output stage AMP.sub.2 is configured to receiveas input signalthe output signal V.sub.O of the input stage AMP.sub.1 and to provideat an output OUTan output voltage VOUT based on the input signal. In most embodiments, the output voltage V.sub.OUT is substantially proportional to the voltage V.sub.O. The output stage AMP.sub.2 is usually a buffer amplifier which is capable of driving capacitive loads. In most implementations, the output stage AMP.sub.2 has a unity gain.

[0020] The inverting input of the input stage AMP.sub.1 is connected to the first input node ISN via a first resistor R.sub.1, and the non-inverting input is connected to the second input node ISP via a second resistor R.sub.2. As mentioned, the current sense resistor R.sub.S is connected between the nodes ISP and ISN. A feed-back path couples the output OUT of the output stage AMP.sub.2 with the inverting input of the input stage AMP.sub.1. In the depicted example, the feed-back path includes only a resistor R.sub.3. The non-inverting input of the input stage AMP.sub.1 is coupled, via resistor R.sub.4, to a circuit node, to which a reference voltage V.sub.REF is supplied. Assuming R.sub.1=R.sub.2=R and R.sub.3=R.sub.4=N. R the gain of the current sense amplifier circuit is N, i.e. V.sub.OUT=N.Math.V.sub.S=i.sub.S.Math.N.Math.R.sub.S.

[0021] In the depicted example, the reference voltage is provided by an operational amplifier OA, which is configured to amplify the voltage provided at the middle tap of a voltage divider composed of resistors R.sub.5 and R.sub.6. The operational amplifier OA is configured as a buffer amplifier, i.e. its output is connected to its inverting input, wherein the non-inverting input is connected to the middle-tap of the mentioned voltage divider that is coupled between supply voltage V.sub.DD and ground GND. In the present example, in which R.sub.6=R and R.sub.5=4.Math.R, the reference voltage V.sub.REF is 0.2 times the supply voltage V.sub.DD, i.e. V.sub.REF=0.2. V.sub.DD. The capacitor C.sub.OUT connected to the output OUT symbolizes the capacitive load impedance, which the current sense amplifier has to drive. It is understood that the purpose of the reference voltage V.sub.REF is merely to determine the DC level of the output voltage V.sub.OUT when the input voltage V.sub.S=i.sub.S.Math.R.sub.S=0 volts.

[0022] As mentioned above, the output stage AMP.sub.2 may be implemented using a source follower in the embodiments described herein. An example is shown in FIG. 2. Accordingly, the output stage AMP.sub.2 includes an n-channel MOS (metal-oxide-semiconductor) field-effect transistor (FET) T.sub.1 whose drain electrode is connected to a supply node (supply voltage V.sub.DD)) and whose source electrode is connected to the output node OUT. The gate electrode of transistor T.sub.1 is connected to the output of the input stage AMP.sub.1 and receivesas input voltagethe voltage V.sub.O. A current source Q.sub.1 is connected between the output node OUT and the ground node GND. The current source Q.sub.1 provides a bias current i.sub.0. Various ways of implementing the current source are as such known to a skilled person and are thus not discussed herein in more detail. It is understood, that FIG. 2 shows merely an example. In other embodiments, the circuit may be flipped, wherein the n-channel MOSFET is replaced by a p-channel MOSFET to obtain a complementary circuit. Furthermore, it is understood that the MOSFET T.sub.1 may be replaced by a bipolar junction transistor (npn- or pnp-type dependent on the actual implementation.

[0023] The concept described herein allows driving capacitive loads (see FIG. 2, capacitance C.sub.OUT) with a low distortion output stage (such as the source follower discussed above) that does not negatively affect the linearity, loop stability and transient response time, while simultaneously keeping the power dissipation low (low quiescent current i.sub.0).

[0024] As can be seen in FIG. 2, source follower output stages are typically asymmetric with respect to the transient response times for charging and discharging the output capacitance C.sub.OUT. Depending on the implementation (i.e. dependent on whether a p-channel MOSFET or an n-channel MOSFET is used) either the transient response time for discharging or for charging the output capacitance can easily be reduced by properly sizing the MOSFET, whereas the other response time is determined by the low quiescent current i.sub.0.

[0025] In the example of FIG. 2 (n-channel MOSFET) the transient time for discharging the output capacitance can be reduced by drastically increasing the bias current i.sub.0, since the slew rate of the output is given by the magnitude of the (static) bias current i.sub.0. However, a massive increase of the bias current i.sub.0 is undesired as this would significantly increase losses. Using a class AB output stage would resolve the problem of asymmetric transient response times but deteriorate linearity for a given open loop gain (which is necessary to achieve the desired loop stability and a low offset). The example shown in FIG. 3 provides a compromise between low losses, high slew rate, and high linearity.

[0026] The example of FIG. 3 is the same as the circuit of FIG. 1 except that the amplifier circuit includes an additional feed-forward circuit FF that is configured to activate a current path coupled to the output OUT to provide additional output current i.sub.dyn, when the differential input voltage V.sub.P?V.sub.N crosses a specific (determined by the circuit design) threshold value V.sub.OS.

[0027] The current path of the feed-forward circuit FF may include a controllable current source coupled to the output of the output stage AMP2. In the depicted example, the controllable current source is implemented using an n-channel MOSFET T.sub.2 whose drain-source current path lies between the output node OUT and ground node GND. When a sufficiently high gate voltage is applied to the gate electrode of the transistor T.sub.2, then the transistor T.sub.2 becomes conductive and an additional bias current i.sub.dyn may flow from the output node OUT to ground (in additional to bias current i.sub.0, see FIG. 2).

[0028] The mentioned controllable current source (e.g. the transistor T.sub.2) is activated, when the differential input voltage V.sub.P?V.sub.N crosses the mentioned threshold value V.sub.OS. In the depicted example, the gate voltage for transistor T.sub.2 is provided by a differential amplifier OA.sub.2, which may also be an operational amplifier. The amplifier OA.sub.2 receivesas input voltagethe voltage V.sub.N?V.sub.P?V.sub.OS, wherein the offset voltage V.sub.OS is provided by a voltage source Q0 coupled between the inverting input of the input stage AMP.sub.1 and the non-inverting input of the differential amplifier OA.sub.2. As a consequence, the differential amplifier OA.sub.2 generates a positive gate voltage when differential input voltage V.sub.P?V.sub.N falls below the (negative) threshold value?V.sub.OS. It is understood that FIG. 3 is merely an example and a skilled person is able to implement basically the same function in a different way. For example, an npn-type bipolar junction transistor may be used instead of MOSFET T.sub.2. Dependent on the implementation of the output stage AMP.sub.2 (source follower), a complementary circuit may be used, in which n-channel transistors are replaced by p-channel transistors while the circuit is flipped.

[0029] The feed-forward circuit FF may be seen as an additional compensation loop, which is able to provide a current-path for a fast discharge of larger capacitive loads when negative differential transients appear at the input of the amplifier circuit. The static current consumption (and thus the losses) is not increased because the compensation loop activates the discharge current-path only temporarily while the differential input voltage V.sub.P-V.sub.N is sufficiently negative (or sufficiently positive if complementary implementations are used for the source follower and the controllable current source (transistor T.sub.2)).

[0030] The concept described herein is enables keeping the DC current consumption low (low bias current i.sub.0 in the output stage) even when using capacitive loads C.sub.OUT of, e.g., 2 nF which is a typical value in motor control applications. At the same time, the concept described herein allows a fast response to large differential voltage swings at the input independent of the load capacitance C.sub.OUT. A response time as low as 1 ?s may be achieved for maximum loads of 2.5 nF. In contrast thereto, in conventional applications the output capacitance range is specified as 10 pF up to 400 pF. Accordingly, the embodiments presented herein may be able to drive an output capacitance more than six times higher than conventional approaches.

[0031] FIG. 4 illustrates one example of a practical implementation of the concept illustrated in FIG. 3. It is noted that only the input stage AMP.sub.1, the output stage AMP.sub.2 and the feed-forward circuit FF are shown in FIG. 4 to keep the illustration simple. The rest of the circuit may be implemented as shown in FIG. 3.

[0032] As shown in FIG. 4, the input stage AMP.sub.1 includes an operational transconductance amplifier OTA with a differential current output (currents i.sub.P and i.sub.N). The output current i.sub.P of the OTA is supplied to a first current mirror composed of the (n-type) transistors T.sub.A and T.sub.B. Similarly, the output current i.sub.N of the OTA is supplied to a second current mirror composed of the (p-type) transistors T.sub.C and T.sub.D. As shown in the depicted example, the transistors T.sub.A, T.sub.B, T.sub.C, and T.sub.D may be field effect transistors. Assuming that the current mirrors have a unity gain, the current mirrors do only invert the direction of the currents without scaling. The output branches of the first and second current mirrors are connected at an output circuit node, at which the output voltage Vo of the input stage AMP.sub.1 is provided. This output voltage V.sub.O is supplied to the input of the output stage AMP.sub.2, i.e. to the gate of transistor T.sub.1 which is configured as source follower as explained above with reference to FIG. 2. Current source Q.sub.1 provides the static bias current for the transistor T.sub.1. Reference is made to the description of FIG. 2 to avoid unnecessary reiterations.

[0033] As explained above, the additional current path (for the current i.sub.dyn) in the feed-forward circuit is activated when differential input voltage V.sub.P?V.sub.N of the input stage AMP.sub.1 falls below the (negative) threshold value ?V.sub.OS. In the previous example of FIG. 3, the condition V.sub.P?V.sub.N<?V.sub.OS is directly evaluated by the differential amplifier OA.sub.2 (and using the offset voltage source Q.sub.0). FIG. 4 illustrates an alternative approach for evaluating the condition V.sub.P?V.sub.N<?V.sub.OS. For this purpose, a replica of the current i.sub.P is generated by adding an additional output branch (transistor T.sub.B) to the first current mirror. As can be seen from FIG. 4, the output current i.sub.P of the OTA in the input stage AMP.sub.1 is drained via transistor T.sub.A which forms the input branch of the first current mirror. The current i.sub.P is copied to the output branch formed by transistor T.sub.B and an additional replica current i.sub.REP=i.sub.P flows through transistor T.sub.B. The source electrode of transistor T.sub.B is coupled to ground GND, while the drain electrode of transistor T.sub.B is coupled to the input branch of a third current mirror composed of (p-type) transistors T.sub.E and T.sub.F. The replica current i.sub.REP=i.sub.P is then again copied by the third (e.g. unity gain) current mirror. The concept of current mirrors with multiple output branches is well known and thus not further discussed herein.

[0034] The offset voltage V.sub.OS is represented by an equivalent offset current i.sub.OS, which is provided by current source Q.sub.OS. The current source Q.sub.OS is connected between a ground node (GND) and the drain electrode of transistor T.sub.F. That is, the current source Q.sub.OS is coupled to the output branch of the third current mirror (transistors T.sub.E and T.sub.F). The offset current i.sub.OS is determined by circuit design and is usually constant. The replica current i.sub.REP=i.sub.P is determined by the output current i.sub.P of the OTA. As a consequence the difference current i.sub.D=i.sub.REP-i.sub.OS has to be drained from the circuit node at which the current source Q.sub.OS and transistor T.sub.F are connected. In the present example, this difference current is drained (to ground) via the input branch of a fourth current mirror that is composed of transistor T.sub.G (input branch) and transistor T.sub.H (output branch). This current mirror may be configured to amplify the current i.sub.D in the input branch. Accordingly, the current i.sub.dyn in the output branch of the third current mirror may be proportional to current i.sub.D (i.e. i.sub.dyn=K.Math.i.sub.D, proportionality factor K). The transistor T.sub.H in the output branch of the third current mirror has the same function as transistor T.sub.2 in the previous example of FIG. 3.

[0035] When comparing the circuits of FIGS. 3 and 4, one can see that, in the example of FIG. 4, only four transistors, namely T.sub.B, T.sub.E, T.sub.F and T.sub.G, are needed instead of the differential amplifier OA.sub.2 while an offset current source Q.sub.OS is used instead of the offset voltage source Q.sub.0. The transistor TH may be seen as a controllable current source because the current i.sub.dyn passing through the transistor T.sub.F is controlled by the current difference i.sub.D. As the output current i.sub.P of the OTA is inversely proportional to the differential input voltage V.sub.P?V.sub.N, the difference current i.sub.D=i.sub.P-i.sub.OS (assuming i.sub.REP=i.sub.P) is indicative of the voltage V.sub.N?V.sub.P?V.sub.OS, which, in FIG. 3, is the differential input of the differential amplifier OA.sub.2. The third current mirror (transistors T.sub.E and T.sub.F) amplify the current i.sub.D to obtain the current i.sub.dyn and thus has basically the same function has the differential amplifier OA.sub.2 in FIG. 3. Accordingly, the feed-forward circuit in the example of FIG. 4 is equivalent to the feed-forward circuit of FIG. 3, wherein, however, the example of FIG. 4 is simpler to implement.

[0036] It is again, emphasized that the circuit of FIG. 4 can be transformed in a complementary circuit, in which the n-type transistors and p-type transistors change roles. For example, in such a complementary circuit, the transistor T.sub.H would be a p-type transistor coupled between the output OUT and the supply node (voltage V.sub.DD). An example is illustrated in FIG. 5. The function is substantially the same as the function of the circuit of FIG. 4 and reference is made to the respective explanations above. Figuratively speaking, the circuit of FIG. 5 is flipped upside down as compared to the circuit of FIG. 4, wherein each p-type transistor is replaced by a complementary n-type transistor and vice versa. Further, in FIG. 5, the replica current i.sub.REP provided by transistor T.sub.B is a replica of the output current i.sub.n of the OTA. The difference current i.sub.P is therefore i.sub.REP?i.sub.OS=i.sub.N?i.sub.OS (in case the current mirror composed of transistors T.sub.A and T.sub.B has unity gain).

[0037] In the example of FIG. 4, as long as the magnitude of differential input voltage V.sub.P?V.sub.N is lower than the offset voltage, the feed-forward circuit has substantially no effect. The DC current consumption of the output stage AMP.sub.2 is low and determined by the bias current source Q.sub.1 (see FIGS. 2 and 4). Only when the differential input voltage V.sub.P?V.sub.N falls below the offset?V.sub.OS (or, in case of the complementary circuit, exceeds the offset V.sub.OS) the feed-forward circuit becomes active and provides a current path (e.g. transistor T.sub.2 or T.sub.H, see FIGS. 3, 4 and 5) for quickly discharging (or, in case of the complementary circuit, for quickly charging) the output capacitance C.sub.OUT.

[0038] Various embodiments described herein are summarized below. It is understood that the following is not an exhaustive list but rather an exemplary summary. A first embodiment relates to an amplifier circuit that includes an input stage and an output stage (see FIGS. 3 and 4, AMP.sub.1 and AMP.sub.2). The input stage has a non-inverting input (+) and an inverting input (?) for receiving a differential input voltage (see FIGS. 3 and 4, voltage V.sub.P?V.sub.N) and is configured to provide an output signal that represents the differential input voltage. The output stage is configured to receiveas input signalthe output signal of the input stage and to provideat an amplifier outputan output voltage based on the input signal. A feed-back path couples the amplifier output with the inverting input of the input stage. In a simple embodiment, the feed-back path basically consists of a resistor (see FIG. 3, feed-back resistor R.sub.3). A feed-forward circuit is configured to activate a current path coupled to the amplifier output to provide additional output current (see FIGS. 3 and 4. current i.sub.dyn) when the differential input voltage crosses (e.g. exceeds or falls below, dependent on the actual implementation) a threshold value (see FIGS. 3 and 4. offset voltage V.sub.OS and, respectively, offset current i.sub.OS).

[0039] One embodiment specifically relates to a current sensing application. In such an embodiment, a current sense resistor is connected between a first input node and a second input node (see FIG. 3, ISN and ISP) The inverting input of the input stage is connected to the first input node via a first resistor and the non-inverting input of the input stage is connected to the second input node via a second resistor. The non-inverting input is connected to a reference voltage source via another resistor (see FIG. 3, reference voltage V.sub.REF).

[0040] The current path of the feed-forward circuit may include a controllable current source coupled to the amplifier output, wherein the controllable current source is activated when the differential input voltage crosses the mentioned threshold value (e.g. determined by V.sub.OS or i.sub.OS, see FIGS. 3 and 4).

[0041] The output stage may include a source follower (see FIG. 2), which provides high linearity and loop-stability. The source follower may be composed of an n-channel field effect transistor, while the feed-forward circuit is configured to sink the additional output current, when the differential input voltage falls below the threshold value (see FIGS. 3 and 4). As mentioned above, the embodiments shown in the Figures may readily be transformed into complementary circuits. In these embodiments the source follower is composed of a p-channel field effect transistor, wherein the feed-forward circuit is configured to source the additional output current when the differential input voltage exceeds the threshold value.

[0042] A further embodiment relates to a method for operating an amplifier. The method includes providingby an input stage of the amplifieran output signal that represents a differential input voltage of the input stage. The method further includes providingby an output stage of the amplifieran output voltage at an amplifier output based on the output signal of the input stage, wherein a feed-back path couples the amplifier output with an inverting input of the input stage. Furthermore, the method includes activating a current path coupled to the amplifier output to provide additional output current, when the differential input voltage crosses a threshold value.

[0043] Although the invention has been illustrated and described with respect to one or more implementations, alterations and/or modifications may be made to the illustrated examples without departing from the spirit and scope of the appended claims. In particular regard to the various functions performed by the above described components or structures (units, assemblies, devices, circuits, systems, etc.), the terms (including a reference to a means) used to describe such components are intended to correspondunless otherwise indicatedto any component or structure, which performs the specified function of the described component (e.g., that is functionally equivalent), even though not structurally equivalent to the disclosed structure, which performs the function in the herein illustrated exemplary implementations of the invention.