Background of the Invention
1. Field Of The Invention
This invention relates to the design of differential amplifiers that are fabricated from Field Effect Transistors (FET) and in particular the factors of the design that relate to the enhancement of the gain of single stage cascoded differential amplifiers.
2. Description of Related Art
Differential amplifiers are incorporated in the design and fabrication of large scale integrated circuits. These differential amplifiers are the essential building blocks for such application as high resolution, high clock rate sample and hold structures and data converters.
A single channel Metal Oxide Semiconductor (MOS) cascaded stages amplifier design as described in "MOS Sampled Data Recursive Filters Using Switched Capacitor Integrators" by Hosticka, et al., IEEE Journal of Solid State Circuits Vol. SC-12, No. 6 Dec. 1977, pp 250-258 is shown in FIG. 1. The input stage is a single ended amplifier formed by n-channel MOS (n-MOS) FET's M1, M2, M3, and M4. The input stage is followed by a level shifter formed by n-MOS FET's M8 and M9 which is driving the main gain stage formed by n-MOS FET's M10, M1, M12, M13, M14, M15, and M16. The transconductance of the drive transistor M10 is enhanced by the adding of the current source M11 and M12 which injects more DC bias current into M10 and increasing the DC bias current for M16 through M15, thus enhancing the total gain for the stage. The n-MOS FET M3, wired in a common gate configuration, isolates the load device M14 from the driver transistor M10. The output stage formed by n-MOS FET's M17 and M18 are connected in a push-pull driver configuration, but the main gain stage, FET's M10, M1, M12, M13, M14, M15, and M16, has so much gain that M18 is acting as a source follower with M17 as a load.
The circuit is designed so as to be immune to threshold voltage variations. The quiescent voltages through the amplifier are made to track the voltages generated by the bias voltage string M5, M6, and M7.
FIG. 1 illustrates a single channel depletion and enhancement device MOS design with cascaded stages and is not suitable for commercially available CMOS process and high frequency application.
U.S. Pat. No. 5,070,306 (issued Dec. 3, 1991 to Okamoto for a "High-Gain Differential Amplifier Circuit Fabricated From Field Effect Transistors") describes a technique for the improvement of the gain of differential amplifiers by the controlling of the transconductance of the load transistors of each side of the differential pair.
U.S. Pat. No. 5,442,318 (issued Aug. 15, 1995 to Badyal, et al. for "Gain Enhancement Technique for Operational Amplifiers") shows a circuit technique for the enhancement of the gain of operational amplifiers by the coupling of the input differential amplifier to a cascoded current source and feeding back of the output common mode noise to the input operational amplifier to improve the common mode noise rejection ratio.
Summary of the Invention
An object of this invention is the creation of a single stage, cascoded differential amplifier of a Complementary Metal Oxide Semiconductor (CMOS) design with enhanced gain for high frequency applications. Furthermore, another object of this invention is the cascoded differential amplifier capable of operation in high gain, high clock rate, sample and hold and data conversion circuits at power supply level below 3 V.
According to the present invention a single stage, cascoded differential amplifier consists of an amplifier section, a current injection section, and a voltage biasing section. The fully differential amplifier section will amplify the difference of the magnitude of a positive and a negative input voltages. The amplifier section consists of a differential pair of n-MOS FET's with their sources coupled together and connected to a current source. The gate of the first n-MOS FET of the differential pair is connected to the positive input voltage and the second n-MOS FET of the differential pair is connected to the negative input voltage. The drains of the pair of n-MOS FET's are connected to a second pair of n-MOS FET's which isolate the first pair of n-MOS FET's from a first pair of p-channel Metal Oxide Semiconductor (p-MOS) FET's that are configured as current source loads of the amplifier section. The second pair of n-MOS FET's and the first pair of p-MOS FET's are connected to a second pair of p-MOS FET's and a third and fourth pair of n-MOS FET's acting as cascode current sources that are configured as a level translation and buffer circuitry that will buffer and isolate a positive output terminal and a negative output terminal from the differential pair of n-MOS FET's and so as to deliver the desired output signal swing at the positive and negative output terminals.
The current injection section is formed by two pairs of p-MOS FET's configured as current sources. These current sources inject current into the drains of the differential pair of the differential amplifier section thus maintaining or increasing the transconductance of the amplifier section.
The voltage biasing section provide the voltages necessary to establish the correct current in the current sources of the amplifier section and the current injection section. The voltage biasing also provides the necessary voltage biases to establish the positive and negative output voltage range at the positive and negative output terminals in the level translation and buffer circuitry of the amplifier section.
Brief Description of the Drawings
FIG. 1 is a schematic diagram of an n-MOS differential amplifier design of prior art having current injection for gain enhancement.
FIG. 2 is a block diagram of the functions of a differential amplifier with current injection for gain enhancement of the present invention.
FIG. 3 is a schematic diagram of the amplifier section of the present invention.
FIG. 4 is a schematic diagram of the current injection section of the present invention.
FIG. 5 is a schematic diagram of the voltage biasing section of the present invention.
Detailed Description of the Invention
Referring to FIG. 2, the amplifier section boosts the magnitude of the difference of the voltages Vinp and Vinn and places this difference on the output terminals Voutp and Voutn. In order to enhance the gain of the amplifier, the current injection section generates currents I.sub.inj and I.sub.inj' to augment the current in the amplifier section. The dc voltage biasing generator enables low voltage operation and establishes the voltages V.sub.a. V.sub.b, V.sub.C. V.sub.d. V.sub.e, and V.sub.f, which adjust the levels of the injection currents I.sub.inj and I.sub.inj', and the levels of the currents within the amplifier section that determine the gain of the amplifier, and the desired signal swing of the Voutp and Voutn. This structure also allows the single stage, cascoded differential amplifier circuit of this invention to operate with voltage supply levels from Vdd to ground of 3.0 V and below.
FIG. 3 is a schematic diagram of the amplifier section of FIG. 2. The n-MOS FET's N.sub.0 and N.sub.2 form a differential pair with their sources connected and coupled to the current source formed by n-MOS FET N.sub.3. The voltage V.sub.f establishes the current level of the current source.
The gate of N.sub.0 is connected to Vinp and the gate of N.sub.2 is connected to Vinn. The n-MOS FET's N.sub.65 and N.sub.66 serve to isolate the differential pair N.sub.0 and N.sub.2 from their current source loads formed by p-MOS FET's P.sub.4 and P.sub.1. This isolation enables the transconductance of the differential pair to be maintained or even increased by the injection of currents I.sub.inj and I.sub.inj' to the drains of N.sub.0 and N.sub.2 respectively. The current injections does not shunt off the original load current of N.sub.0 and N.sub.2 from the current source loads P.sub.4 and P.sub.1 respectively, since the isolation provided by N.sub.65 and N.sub.66 create a low impedance node at the drain of N.sub.0 and N.sub.2. The gate bias for N.sub.65 and N.sub.66 to provide the isolation is established by the bias voltage V.sub.c.
By utilizing the injected currents I.sub.inj and I.sub.inj' to augment the transconductance of N.sub.0 and N.sub.2, less current needs to be drawn from the current source loads P.sub.4 and P.sub.1. This lowers the output conductance of P.sub.4 and P.sub.1, thus enhancing the overall gain of the single stage amplifier section.
The voltage V.sub.a is used to established the current levels within the current source loads P.sub.4 and P.sub.1. The p-MOS FET's P.sub.5 and P.sub.6 with the n-MOS FET's N.sub.26, N.sub.27, N.sub.7, and N.sub.8 form a network to buffer and translate the output voltages Voutn and Voutp from the input differential pair. The appropriate operating bias for the p-MOS FET's P.sub.5 and P.sub.6 and n-MOS FET's N.sub.26, N.sub.27, N.sub.7, and N.sub.8 are established by the voltages V.sub.b, V.sub.d, and V.sub.e which determines the output signal swing at V.sub.outn and V.sub.outp.
The generation of the injected currents I.sub.INJ and I.sub.INJ' is shown in FIG. 4. The p-MOS FET's P.sub.69 and P.sub.70 form a current source with their sources connected to the voltage power supply V.sub.DD. The p-MOS FET's P.sub.71 and P.sub.72 are the cascoded respectively with p-MOS FET's P.sub.70 and P.sub.69. The bias voltages V.sub.a and V.sub.b establish the level of the currents I.sub.INJ and I.sub.INJ'.
The DC Voltage Biasing generator is shown in FIG. 5. Current I.sub.b passes through n-MOS FET N.sub.33 which is connected in a diode configuration. The voltage developed at the drain and gate to source forms the voltage V.sub.f. This voltage biases n-MOS FET's N.sub.34 and N.sub.39 to conduct with currents that are proportional to the current I.sub.b. The current through N.sub.34 serves as the input current for the diode connected p-MOS FET P.sub.35. This establishes the voltage bias V.sub.b and serves to bias p-MOS FET P.sub.57. The voltage bias V.sub.b also biases p-MOS FET P.sub.56 which cascodes the current source p-MOS FET P.sub.38 whose current is in proportion to the current through P.sub.40. The current through P.sub.40 and P.sub.57 and n-MOS FET N.sub.61 is driven by the current of n-MOS FET N.sub.39. This resulted in the V.sub.c developed at the drain of P.sub.57 and the gate of P.sub.40. The p-MOS FET P.sub.36 is also biased by voltage V.sub.b to conduct at a current level that is proportional to the current through P.sub.35. This current develops the voltage V.sub.d across the diode connected n-MOS FET N.sub.37. The voltage V.sub.d forms the bias for the n-MOS FET N.sub.61 which forms a cascoded current source with n-MOS FET N.sub.39. The voltage V.sub.d also forms the bias for the n-MOS FET N.sub.60 which forms a cascoded diode connected configuration with n-MOS FET N.sub.41. The current through N.sub.60 and N.sub.41 is established by the cascoded current source P.sub.38 and P.sub.56. This establishes voltage V.sub.e at the drain of P.sub.56 and the drain of N.sub.60.
The circuit this invention has been described using n-channel and p-channel MOS FET devices. It will be obvious to those skilled in the art that with appropriate changes in materials in the fabrication of this invention that the channel types may be reversed. Also, it will be apparent to those skilled in the art that this invention may be implemented as bipolar junction transistor. These and other changes in form and detail may be made without departing from the spirit and scope of the invention.