This invention relates to an improvement of an amplifier that is useful particularly when it is used for high-frequency amplification applications.
In a transistor circuit that amplifies a high-frequency signal, a phenomenon occurs such that the input capacitance of a base input circuit increases because of the influence of the electrostatic capacitance between a collector and a base. This phenomenon is called Miller effect.
As the frequency to be amplified becomes higher, this phenomenon becomes harder to ignore. Also because of the increase of a noise signal that occurs after a while, an amplification effect cannot be achieved. Thus, conventionally, various techniques for improving the characteristics of a high-frequency amplifier including the Miller effect have been proposed.
For example, in Patent Reference 1, in order to restrain changes of the passing phase of power even when input power increases to cause operation in a nonlinear operating area, PN junction of a diode or transistor is connected in a forward direction to a circuit that supplies bias to a base terminal of a high-frequency amplification transistor of ground amplification type, and a capacitor with small impedance is inserted between its anode and ground.
The detailed description of this circuit is given in Patent Reference 1. By causing a change in the capacitance between the base and emitter of the amplification transistor due to a change of input signal, and a change in the capacitance of the diode or transistor connected to the outside of the base, to increase or decrease in the ways opposite to each other, the apparent change in the input capacitance of the base electrode is reduced.
The circuit using the diode, disclosed in Patent Reference 1, is hereinafter referred to as diode-feed type circuit.
Patent Reference 1:
JP-A-9-260964,
An amplifier according to this invention includes:
an NPN-type first transistor (Q1) having an emitter electrode grounded and having a collector electrode connected to a current control circuit (P1, P2);
an NPN-type second transistor (Q2) having an emitter grounded, the second transistor amplifying a signal inputted to its base electrode and outputting the amplified signal to its collector electrode;
an NPN-type fourth transistor (Q4) having an emitter electrode connected to a base electrode of the first transistor, having a base electrode connected to the collector electrode of the first transistor, and having a collector electrode connected to a power line; and
an NPN-type fifth transistor (Q5) having an emitter electrode connected to the base electrode of the second transistor, having a collector electrode connected to the power line, and having a base electrode connected to the base electrode of the fourth transistor.
With the above-described structure, since the fourth and fifth transistors have both a β difference compensation function and a diode function for diode feed, this amplifier can stably operate at an extremely low voltage.
Embodiment 1
The PFETs (P1, P2) having the current mirror structure are used in a saturation region where a drain current Id is constant. The relation between a drain-source voltage Vds of the PFET and the drain current Id is as shown in
Base-emitter voltages Vbe of a transistor Q3 and a transistor Q1, and an anode-cathode forward voltage Vd of a diode D1 is approximately 0.8 V when the circuit is used under typical conditions. However, it may exceed 0.9 V because of changes in temperature and process variance (difference between individual circuits generated in the manufacturing process).
As is clear from
Vcc>2Vbe+Vd+Vds=3.0(V) (3)
Therefore, as a certain margin is considered so as to prevent any failure to satisfy the expression (3) because of temperature and process variance, the power-supply voltage to be used must be sufficiently higher than 3 V.
If the power-supply voltage is insufficient, for example, a sufficient drain-source voltage Vds of the PFET (P1) of the current mirror that supplies a current to the bias circuit cannot be secured and a desired current cannot be supplied. The entire circuit cannot perform a desired operation.
The transistors Q1, Q2 and Q3 are referred to as NPN-type first, second and third transistors, respectively. The diodes D1 and D2 are referred to as first and second diodes, respectively. The circuit of the PFETs (P1, P2) is referred to as current control circuit.
Embodiment 2
An amplifier according to Embodiment 2 of this invention is shown in
In
To facilitate understanding of the operation of the circuit shown in
In the case of
Ix=M*{1−(M+1)/(β+M+1)}*Iy (1)
In the case of
Ix=M*{1−(M+1)/(β2+β+M+1)}*Iy (2)
Here, β represents the current gain of the transistor. Usually, the value of β largely varies within a range of approximately 150 to 250 because of changes in temperature and process variance generated in the semiconductor manufacturing process. Iy represents a reference current of the current mirror circuit.
As can be understood from the equations (1) and (2), the circuit of
The contribution of β difference of the bias circuit to Ix in
According to the circuit of
Vcc>2Vbe+Vds=2.1(V) (4)
In the circuit of
The transistors Q1, Q2, Q4 and Q5 are referred to as NPN-type first, second, fourth and fifth transistors, respectively.
This invention can be applied not only to a low-noise amplifier (LNA) but also all the circuits that handle particularly high frequencies and need high saturation characteristics, for example, a high-power amplifier or driver amplifier.
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/JP03/03675 | 3/26/2003 | WO | 8/18/2005 |