Field of the Invention
The present invention relates to the field of current-mode feedback operational amplifiers and in particular to current-mode feedback operational amplifiers having low impedance, inverting inputs.
Background of the Invention
The closed loop bandwidth for a conventional current feedback operational amplifier is given by:
Here, C.sub.c is the compensation capacitance at the high gain node of the operational amplifier, R.sub.2 is the feedback resistor, and G is the closed loop voltage gain for the amplifier, which is equal to (1+R.sub.2 /R.sub.1), where R.sub.1 is the gain-setting resistor between the inverting input of the operational amplifier and ground.
For an ideal current feedback operational amplifier, the input impedance, R.sub.in, is zero, and consequently, the closed loop bandwidth of the circuit is independent of the amplifier gain. In conventional current feedback operational amplifiers, however, R.sub.in is non-zero, its magnitude being determined by the intrinsic emitter resistance of the transistors which form the inverting input of the amplifier. This emitter resistance depends on the bias current in the transistor, and for a bipolar transistor it is typically on the order 26 ohms per 1 mA of emitter current.
The feedback resistor, R.sub.2, is typically on the order of a few thousand ohms, and as long as the product, G R.sub.in, is small relative to R.sub.2, the closed loop bandwidth of the amplifier will be virtually independent of gain. However, since R.sub.in is non-zero, G R.sub.in will be comparable R.sub.2 for sufficiently high gains, causing a roll off in B.sub.W.
The transfer function of a closed loop current-feedback operational amplifier and the DC closed-loop gain accuracy, A.sub.c also roll off for large gains. The transfer function is given by:
where Z.sub.t is the open loop transimpedance gain. As with B.sub.W, the transfer function will begin to decrease as the product G R.sub.in becomes comparable to R.sub.2 at high gains. Similarly, the DC closed-loop gain is given by:
where Z.sub.o is the DC open loop transimpedance gain. As with B.sub.W and the transfer function, the gain accuracy falls off for large gains due to its inverse dependence on the factor G R.sub.in.
In addition to these problems, the inverse dependence of R.sub.in on the transistor bias current introduces non-linearities into the circuit response for large input signals. While the effect of such non-linearities can be partially offset by operating the amplifier at higher bias currents, this approach increases power dissipation and noise in the amplifier circuit.
Summary of the Invention
The present invention is an input stage for a current feedback operational amplifier which provides a low impedance inverting input, and consequently reduces the gain dependence of the bandwidth, transfer function, and DC closed-loop accuracy of the circuit. In addition, the bias current of the input stage is amplified at the output stage, providing higher slew rates without increasing the DC power dissipation or noise in the voltage buffer.
The present invention utilizes a new dual output current mirror to reduce the impedance at the inverting input of the voltage buffer by means of a current feedback loop. The dual current mirror combines a Wilson-type current mirror and a transistor current source. The input of the dual current-mirror is driven by the collector of the inverting input transistor. One of the outputs from the dual current-mirror is connected back to the inverting input to form a local current-feedback loop, reducing the input impedance by a factor related to the current gain of the current-mirror. The second output forms the high impedance z-node of the current mirror. Current gain in the new dual current-mirror is determined by the ratio of emitter degeneration resistors.
The current scaling provided by the dual output current mirror has the added advantage of generating a larger current at the high gain node of the amplifier with a smaller biasing current in the voltage buffer. Thus, the output node can operate with sufficient current to insure high slew rates at the amplifier output, while the voltage buffer can operate at lower bias currents, reducing the noise and power dissipation, and increasing the low input impedance associated with high bias currents.
Brief Description of the Drawings
FIG. 1 is a schematic diagram of a conventional current feedback operational amplifier.
FIG. 2 is a schematic diagram of a current feedback operational amplifier including an input stage in accordance with the present invention.
FIG. 3 is a schematic diagram of a dual output current mirror in accordance with the present invention.
FIG. 4 is a detailed schematic diagram of a current feedback operational amplifier including an input stage in accordance with the present invention.
FIG. 5A is a simulation of the bandwidths at a series of different gains for a conventional current feedback operational amplifier.
FIG. 5B is a simulation of the bandwidths at a series of different gains for a current feedback operational amplifier in accordance with the present invention.
FIG. 6 is a schematic diagram of current feedback operational amplifier including an input stage in accordance with the present invention, based on field effect transistors (FETs).
Detailed Description of the Invention
Referring to FIG. 1, there is shown a schematic diagram of a conventional current feedback operational amplifier. A voltage buffer 10 comprises a first pair transistors 12, 14 having complimentary conductivity properties and a second pair of transistors 16, 18 also having complimentary conductivity properties. The bases of transistors 12, 14 are connected together, forming a non-inverting input 20, and the emitters of transistors 16, 18 are connected together, forming an inverting input 22. Current sources 24, 26 provide bias current to transistors 12, 14, 16, 18. Current mirrors 28, 30 provide a replica of that current to point A, the high gain node of the current feedback operational amplifier 40. A feedback circuit 34 comprises a feedback resistor 36 and a gain setting resistor 38 connected as a voltage divider between output 42 of output buffer 32 and inverting input 22.
The open-loop input impedance at inverting input 22 is determined by the parallel resistance of the transistors 16, 18 which is just the intrinsic emitter resistance, R.sub.e, of these devices. Typically, R.sub.e for a bipolar transistor is approximately 25/I.sub.E where I.sub.E is the bias emitter current in milliamps. For example, with an I.sub.E of 1 mA, R.sub.e is approximately 25 ohms, and the input impedance R.sub.IN of inverting input 22 is approximately 12 ohms. For typical feedback resistor values of about 1000 ohms, the product G R.sub.IN becomes comparable to R.sub.2 for gains of about 100 or 40 decibels. However, the bandwidth will begin to roll off well before G reaches 40 dB.
Referring now to FIG. 2, there is shown a diagram of a current feedback operational amplifier 100 having an input stage in accordance with the present invention. Where the circuit elements are the same as in current feedback operational amplifier 40 of FIG. 1, the numbering has been retained. Thus, an input voltage buffer 10 comprises a first pair of complimentary transistors 12, 14 and a second pair of complimentary transistors 16, 18. The bases of transistors 12, 14 are connected together, forming a non-inverting input 20. The collectors of transistors 12, 14 are connected to supply voltages V.sub.EE and V.sub.cc, respectively, and their emitters are connected to current sources 24, 26, respectively. Similarly, the emitters of transistors 16, 18 are connected together, forming an inverting input 22, their bases are connected to the emitters of transistors 12, 14, respectively, and their collectors are connected to current mirrors 50, 150, respectively.
Current mirrors 50, 150 are dual output current mirrors, each having a low impedance input current node 52 and 152, respectively, and a pair of high impedance current nodes 54, 56 and 154, 156, respectively. In current mirror 50, low impedance current node 52 is linked to inverting input 22 through transistor 18, while high impedance current node 54 is connected directly to inverting input 22, and high impedance current node 56 is connected to high gain node, A, of the operational amplifier. The low impedance current node 152 and high impedance current nodes 154 of dual output current mirror 150 are connected to inverting input 22 in a manner corresponding to that of current mirror 50. Similarly, high impedance node 156 is connected to high gain node A.
Referring now to FIG. 3, there is shown a detailed schematic diagram of a dual output current mirror 50 in accordance with the present invention. A transistor 60 has its collector connected to the collector of transistor 18 of inverting input 22, its emitter connected to supply voltage V.sub.EE through a current scaling resistor 62, and its base connected to the bases of transistors 64, 66. The collector of transistor 60 is connected to the base of transistor 74 and forms low impedance input 52 of dual current-mirror 50. Input 52 is driven by the collector current of transistor 18. Transistors 64, 66, 68 are connected as a Wilson-type current mirror 80. The emitters of transistors 64, 66 are each connected to supply voltage V.sub.EE through current scaling resistors 70, 72, respectively, and their bases are connected to each other and to the base of transistor 60. The base and emitter of transistor 68 are connected to the collectors of transistors 64 and 66, respectively. The collector of transistor 68, which is the high impedance output node of Wilson-type current mirror 80, forms the high gain node of operational amplifier input stage 100 with the high impedance current node 156 of dual output current mirror 150. Transistor 74 has its base connected to the collector of transistor 60, its collector connected to inverting input 22, and its emitter connected to the collector of transistor 64. The collector of transistor 74 forms high impedance current node 54 and isolates transistor 64 from inverting input 22.
With dual output mirror 50 configured as above, currents in transistors 74, 68 are determined by the relative sizes of current scaling resistors 62, 70, and 72. For example, where the ratio of resistor 62 to resistor 70 is 1/K:1, currents KI and I are generated at the collectors of transistors 60, 74 which correspond to current nodes 52, 54, respectively, of dual output current mirror 50. Thus, a signal at inverting input 22 which would generate current KI at the inverting input 22 of a conventional current feedback operational amplifier 40, generates a current (1+K)I at inverting input 22 of operational amplifier 100. Accordingly, the impedance of inverting input 22 with feedback from current mirror 50 is given by:
where R.sub.IN is the input impedance of the conventional operational amplifier 40 and R'.sub.IN is the input impedance of operational amplifier 100 with current feedback. By choosing resistors 70 so that K<1, the input impedance of inverting input 22 can be reduced significantly. Further, by choosing resistor 72 to have the same resistance as resistor 70, the current at high impedance node 56 will be amplified over the bias current in voltage buffer 10 by a factor 1/K. Thus, the bias current in voltage buffer 10 can be kept low without limiting the current and, consequently the slew rate, at high gain node A.
Referring now to FIG. 4, there is shown a detailed schematic diagram of a current feedback operational amplifier circuit in accordance with the present invention, including an output voltage buffer 140. Output buffer 140 comprises a pair of complimentary transistors 149, 144, the bases of which are connected to high impedance nodes 56, 156 of current mirrors 50, 150, respectively. A second pair of complimentary transistors 146, 148 have their emitters connected to an output 160 of the feedback amplifier through resistors 162, 164.
Voltage buffer 10 of the operational amplifier input stage 100 includes diode connected NPN and PNP transistors 122, 124, respectively, connected in series with voltage buffer transistors 12, 14, and diode connected PNP and NPN transistors 126, 128, respectively, connected in series with transistor pair 16, 18. Diode connected transistors 122, 126 and 124, 128 are included to offset any voltage differences between the PNP and NPN transistors 12, 16 and 14, 18, respectively, which would otherwise introduce differences between the voltages at non-inverting input 20 and inverting input 22.
Referring now to FIG. 5A, there are shown HSPICE simulations of the bandwidth of a conventional operational amplifier input stage at a series of different closed-loop gains. In FIG. 5B there are shown simulations of the bandwidth at a series of different gains for an operational amplifier having an input stage as shown in FIG. 4. Both simulations were generated using C.sub.c equal to 0.1 pF and R.sub.2 equal to 1000 ohms. For gains of 6 dB, B.sub.W is approximately 200 MHz for both circuit designs. However, for gains of about 40 dB, B.sub.W for the conventional input stage is about a factor of twenty less than its value at 6 dB, while B.sub.W for a low R.sub.IN input stage of the present invention has fallen off by only a factor of five relative to its value at 6 dB. Thus, the low R.sub.IN of the present invention is effective to reduce the gain dependence of the operational amplifier circuit by increasing the gains necessary to make the factor G R.sub. IN comparable to the feedback resistance, R.sub.2. The gain dependence of the transfer function is reduced and the DC closed-loop gain accuracy is improved by the same mechanism.
Referring now to FIG. 6 there is shown a low input resistance current feedback operational amplifier in accordance with the present invention in which bipolar transistors are replaced by FET transistors. FET transistors are indicated by primed reference numbers corresponding to the reference number of the comparably connected bipolar transistor.