EP0900476A2

EP0900476A2 - Voltage-to-current converter with error correction

Info

Publication number: EP0900476A2
Application number: EP98903226A
Authority: EP
Inventors: Evert Seevinck
Original assignee: Koninklijke Philips Electronics NV
Current assignee: Koninklijke Philips NV
Priority date: 1997-03-13
Filing date: 1998-02-27
Publication date: 1999-03-10
Also published as: WO1998040967A2; WO1998040967A3

Abstract

Voltage-current converter including a signal voltage source (2), a series resistor (4) and an electronic circuit with a first transconductance amplifier (12) and a second transconductance amplifier (22). The corresponding inputs (16, 20; 24, 26) of the transconductance amplifiers (12, 22) are connected to receive a voltage difference (VA) between a node (18) and a reference terminal (8). The output (14) of the first transconductance amplifier (12) is connected to an input terminal (6) to which the series resistor (4) is connected as well. The output (25) of the second transconductance amplifier (22) is connected to an output terminal (10). Due to feedback around the first transconductance amplifier (12) an input current (i1) can flow into the input terminal (6). The input current (i1) is amplified or copied by the second transconductance amplifier (22) and is available at an output terminal (10). The input current (i1) is not only dependent on the series resistor (4), but also on the voltage difference (VA) between the node (18) and the reference terminal (8). The distortion caused by the voltage difference is cancelled by measuring (28) this voltage difference and adding (30) this voltage difference to the input voltage signal (VS) in the proper sense.

Description

Voltage-to-current converter with error correction.

The invention relates to an electronic circuit comprising: an input terminal and a reference terminal for receiving an input current flowing through the input teπriinal and the reference terminal; an output terminal for generating an output current flowing through the output terminal and the reference terminal; a first transconductance amplifier having an output coupled to the input terminal and having inputs connected to receive a voltage difference between a first node and the reference node; a second transconductance amplifier having an output coupled to the output terminal and having inputs connected to receive said voltage difference between the first node and the reference node; and means for coupling the input terminal to the first node.

Well-known methods of voltage-current conversion suitable have the disadvantage of signal distortion due to circuit non-linearity, particularly at high frequencies. Figure 1 shows a known general technique. An input voltage V_s from source 2 is connected, via a series resistor 4, to an input terminal 6 and a reference terminal 8 of an electronic circuit which provides an output current i₂ at an output terminal 10 in response to an input current i_j which flows into the input terminal 6. The electronic circuit comprises a first transconductance amplifier 12 with an output 14 coupled to the input terminal 6, an inverting input 16 coupled to a first node 18 and a non-inverting input 20 coupled to the reference terminal 8 which serves as signal ground. A short-circuit between the input terminal 6 and the first node 18 provides a zero signal difference between the input terminal 6 and the first node 18. The electronic circuit further comprises a second transconductance amplifier 22 with an output 25 coupled to the output terminal 10, an inverting input 24 coupled to the first node 18 and a non-inverting input 26 coupled to the reference terminal 8. The first and second transconductance amplifiers 12 and 22 have a transconductance g₂ and g₂, respectively. They can be very simple, e.g. single bipolar or MOS transistors, in which case the output, inverting input and non-inverting input of the transconductance amplifier corresponds to the collector, base and emitter, respectively, of a bipolar transistor or to the drain, gate and source, respectively, of a MOS transistor. Such a two transistor arrangement is known as a current mirror with a current gain i₂/i₁=g₂/gι_> which in combination with the series resistor 4 provides fast voltage-current converter useful up to high frequencies.

However, conversion errors occur. It follows from the circuit diagram that:

V<

J.1 = - ± - ^9l i

R<!+ — 9\

9\

V_A is the signal voltage difference between the first node 18 and the reference terminal 8 and R_s is the resistance of series resistor 4. In the expression for i_1? the transconductance g₁ is normally not well-defined, temperature dependent and nonlinear. Therefore the current i_j will suffer from distortion. The ratio g₂/gι enables variable gain to be achieved and is generally well-defined and linear since the peculiarities of transconductance g₁ are compensated by those of transconductance g₂. It follows that also the output current i₂ will suffer from distortion. It is an object of the invention to provide an electronic circuit arrangement with less distortion. To this end the electronic circuit as specified in the opening paragraph is characterized in that the means for coupling comprises: means for measuring the voltage difference between the first node and the reference terminal; and means for inserting said voltage difference between the input terminal and the first node.

The invention is based on the recognition that the source of distortion is the signal voltage V_A at the first node 18. This error signal is measured and added to the input voltage signal V_s by inserting the measured error voltage between the input terminal 6 and the first node 18. This is shown in Figure 2. Now the current i_j is given by:

i = ^VS^~VA^+VA ₌ VS ^Rs ^Rs

When compared to the above expression it can be seen that the distortion- inducing term has disappeared and the voltage-current conversion is perfect. The technique outlined above may be implemented through the use of two additional transconductance amplifiers, preferably with mutually equal transconductances. To this end in an embodiment of the electronic circuit the means for measuring comprises a third transconductance amplifier having an output coupled to the first node and having inputs connected to receive said voltage difference between the first node and the reference node; and the means for inserting comprises a fourth transconductance amplifier having an output coupled to the first node and having inputs connected to receive a voltage difference between the input terminal and the first node.

The signals at the inputs of the third transconductance amplifier and the fourth transconductance amplifier are reversed copies of each other since the output currents of both transconductance amplifiers are equal but opposite in direction. So, the error signal at the first node appears in opposite direction between the input terminal and the first node, thereby counter-acting the effect of the error signal.

Advantageous embodiments are defined in the dependent claims which specify transistor implementations of the first to fourth transconductance amplifiers. The transistors may be bipolar transistors or unipolar (MOS) transistors.

These and other aspects of the invention will be described in more detail, by way of example, with reference to the accompanying drawings, in which:

Figure 1 shows a circuit diagram of a conventional voltage-current converter;

Figure 2 shows a circuit diagram of a voltage-current converter according to the invention; Figure 3 shows a circuit for elucidating the operation of the voltage- current converter according to the invention.

Figure 4 shows a circuit diagram of an embodiment of an voltage-current converter according to the invention;

Figure 5 shows a circuit diagram of an embodiment of an voltage-current converter according to the invention; and

Figure 6 shows a circuit diagram of an embodiment of an voltage-current converter according to the invention.

In the Figures corresponding elements have the same reference signs. Although the embodiments to be described and shown in the accompanying Figures are illustrated with bipolar transistors, the embodiments can also be implemented with unipolar or MOS transistors. For bipolar transistors the control electrode, first main electrode and second main electrode correspond to the base, emitter and collector, respectively. For MOS transistors the control electrode, first main electrode and second main electrode correspond with the gate, source, and drain, respectively.

Figure 1 shows a known general technique for converting an input voltage V_s into an output current i₂. The input voltage V_s from a source 2 is connected, via a series resistor 4, to an input terminal 6 and a reference terminal 8 of an electronic circuit which provides an output current i₂ at an output terminal 10 in response to an input current i_x which flows into the input terminal 6. The electronic circuit comprises a first transconductance amplifier 12 with an output 14 coupled to the input terminal 6, an inverting input 16 coupled to a first node 18 and a non-inverting input 20 coupled to the reference terminal 8 which serves as signal ground. A short-circuit between the input terminal 6 and the first node 18 provides a zero signal difference between the input terminal 6 and the first node 18. The electronic circuit further comprises a second transconductance amplifier 22 with an output 25 coupled to the output terminal 10, an inverting input 24 coupled to the first node 18 and a non-inverting input 26 coupled to the reference terminal 8. The first and second transconductance amplifiers 12 and 22 have a transconductance gγ and g₂, respectively. They can be very simple, e.g. single bipolar or MOS transistors, in which case the output, inverting input and non-inverting input of the transconductance amplifier corresponds to the collector, base and emitter, respectively, of a bipolar transistor or to the drain, gate and source, respectively, of a MOS transistor. Such a two transistor arrangement is known as a current mirror with a current gain i₂/iι =g₂/gι , which in combination with the series resistor 4 provides fast voltage-current converter useful up to high frequencies. It follows from the circuit diagram that:

^VS^'VA -~~-^ da) ^κs

i₂=-h (Id)

£1

V_A is the signal voltage difference between the first node 18 and the reference terminal 8 and R_s is the resistance of series resistor 4. In expression lc for i_l5 the transconductance g_l is normally not well-defined, temperature dependent and nonlinear. Therefore the current i_j will suffer from distortion. The ratio g₂/g enables variable gain to be achieved and is generally well-defined and linear since the peculiarities of transconductance g_x are compensated by those of transconductance g₂. It follows that also the output current i₂ will suffer from distortion. Figure 2 shows a circuit diagram of a voltage-current converter according to the invention. The technique employed is based on the recognition that the source of distortion is the signal voltage V_A at the first node 18. This error signal V_A is measured, symbolically depicted with sensing or measuring means 28, and added to the input voltage signal V_s by inserting the measured error voltage between the input terminal 6 and the first node 18, symbolically depicted with voltage source 30. In this configuration the current i_t is given by: j_{1 =} ^VS S~^VA A^+VA A ₌_ ^vsS ₍₂₎

^Rs ^Rs

When compared to expression lc it can be seen that the distortion- inducing term V_A has disappeared and the voltage-current conversion i₂/V_s is perfect. The compensation technique involves positive feedback combined with the negative feedback already present around transconductance amplifier 12. The stability is considered in connection with the circuit diagram of Figure 3 in which the loop around transconductance amplifier 12 is opened at the nodes 18 and 18a. A signal V; is applied to node 18 and the resulting signal at node 18a is V₀. The open-loop gain, i.e. V₀/V_j the follows from the following equations: ^vo^=vA- ^Rs (3a)

*ι-*ιV, (3b) o

V_A-V_t (3c)

^-1-*Λ (3e) i

For closed-loop stability the open-loop gain should be < 1. From equation 3e it follows that: _8lR_s>0 Of)

This condition is always fulfilled since both g_x and R_s are positive.

Figure 4 shows an implementation of the error correction technique through the use of two additional transconductance amplifiers. The error voltage V_A is measured by a third transconductance amplifier 32, which has a non-inverting input 34 connected to the node 18, an inverting input 36 connected to the reference terminal 8 and an output 38 connected to the node 18. An opposite error voltage V_A is inserted between the input terminal 6 and the node 18 by a fourth transconductance amplifier 40, which has a non- inverting input 42 connected to the input terminal 6, and an inverting input 44 and an output 45 connected to the node 18. The output current from the output 38 of transconductance amplifier 32 flows into the output 46 of transconductance amplifier 40 and causes at the inputs 42 and 44 of transconductance amplifier 40 a voltage with an amplitude equal to the amplitude of the error voltage V_A, but with opposite sign.

All four transconductance amplifiers 12, 22, 32 and 40 can be very simple. The third transconductance amplifier 32 and the fourth transconductance amplifier 40 preferably have substantially equal transconductances g₀ in order to achieve the best error correction performance. The transconductances g_l and g₂ of the other two transconductance amplifiers 12 and 22 can be unequal in order to obtain current gain. The overall transfer function of the voltage-current converter is given by:

Si = (4) Si R<

This is an accurate, distortionless transfer with variable gain g₂/g_1? and valid up to high frequencies.

Figure 5 shows a first transistor implementation of the four transconductance amplifier configuration of Figure 4. The first transconductance amplifier 12 is implemented with transistor Tl, which has its emitter, base and collector connected to the reference terminal 8, the first node 18 and the input terminal 6, respectively. The second transconductance amplifier 22 is implemented with transistor T2, which has its emitter, base and collector connected to the reference terminal 8, the first node 18 and a second node 48, respectively. Transistor T3 and transistor T4 form the third transconductance amplifier 32. The emitter, base and collector of transistor T3 are connected to the second node 48, a bias voltage terminal 50 and the output terminal 10, respectively. The collector of transistor T2, which corresponds to the output 25 of the second transconductance amplifier 22, is thus coupled to the output terminal 10 via the main current path of transistor T3. The emitter, base and collector of transistor T4 are connected to the reference terminal 8, the second node 48 and the first node 18, respectively. Transistor T5 is the fourth transconductance amplifier 40 and has its emitter, base and collector connected to the first node 18, the input terminal 6 and a supply voltage terminal 52. Coupling the emitter current of transistor T5 to the first node 18 has substantially the same effect as coupling the collector current to the first node 18, as would be expected from Figure 4, because the emitter current and collector current if transistor T5 are nearly equal.

The bias voltage terminal 50 provides a bias voltage V_B generated by means of two diode-connected transistors T7 and T8 series connected between the bias voltage terminal 50 and the reference terminal 8 and a bias current source 54 connected between the bias voltage terminal 50 and the supply voltage terminal 52. The base-emitter voltage of transistor T2 is equal to V_A. Since the current through transistor T2 flows through transistor T3, the base-emitter voltage of transistor T3 is also V_A. This implies that the voltage at the second node 48 and also the base-emitter voltage of transistor T4 is equal to V_B-V_A. The voltage V_B-V_A in turn appears at the base-emitter junction of transistor T5 since the currents through transistors T4 and T5 are substantially equal. It appears that the voltage at the input terminal 6 is fixed at the bias voltage V_B and no longer dependent on the error voltage V_A at the first node 18.

Figure 6 shows a second transistor implementation which differs from the implementation shown in Figure 5 in that the output current i₂ is not taken from the collector of transistor T3, but from the collector of a further transistor T6 which has its emitter, base and collector connected to the reference terminal 8, the first node 18 and the output terminal 10, respectively. The collector of transistor T3 is connected to the supply terminal 52. The current through transistor T2 is copied or multiplied in transistor T6. The additional transistor T6 makes this circuit suitable for use at high frequencies.

Claims

CLAIMS:

1. An electronic circuit comprising: an input terminal (6) and a reference terminal (8) for receiving an input current flowing through the input terminal (6) and the reference terminal (8); an output terminal (10) for generating an output current flowing through the output terminal (10) and the reference terminal (8); a first transconductance amplifier (12) having an output (14) coupled to the input terminal (6) and having inputs (16, 20) connected to receive a voltage difference between a first node (18) and the reference node (8); a second transconductance amplifier (22) having an output (25) coupled to the output terminal (10) and having inputs (24, 26) connected to receive said voltage difference between the first node (18) and the reference node (8); and means for coupling the input terminal (6) to the first node (18), characterized in that the means for coupling comprises: means (28) for measuring the voltage difference between the first node (18) and the reference terminal (8); and means (30) for inserting said voltage difference between the input terminal (6) and the first node (18).

2. An electronic circuit as claimed in claim 1, wherein the means (28) for measuring comprises a third transconductance amplifier (32) having an output (38) coupled to the first node (18) and having inputs (34, 36) connected to receive said voltage difference between the first node (18) and the reference node (8); and the means (30) for inserting comprises a fourth transconductance amplifier (40) having an output (46) coupled to the first node (18) and having inputs (42) connected to receive a voltage difference between the input terminal (6) and the first node (18).

3. An electronic circuit as claimed in claim 2, wherein: the first transconductance amplifier (12) comprises a first transistor (Tl) having a first main electrode coupled to the reference terminal (8), a second main electrode coupled to the input terminal (6) and a control electrode coupled to the first node (18); the third transconductance amplifier (32) comprises a second transistor (T3) having a first main electrode coupled to a second node (48), a second main electrode coupled to the output terminal (10) and a control electrode coupled to a bias voltage terminal (50), and a third transistor (T4) having a first main electrode coupled to the reference terminal (8), a second main electrode coupled to the first node (18) and a control electrode coupled to the second node (48); the second transconductance amplifier (22) comprises a fourth transistor (T2) having a first main electrode coupled to the reference terminal (8), a second main electrode coupled to the second node (48) and a control electrode coupled to the first node (18); and the fourth transconductance amplifier (40) comprises a fifth transistor (T5) having a first main electrode coupled to the first node (18), a second main electrode coupled to a supply voltage terminal (52) and a control electrode coupled to the input terminal (6).

4. An electronic circuit as claimed in claim 2, wherein: the first transconductance amplifier (12) comprises a first transistor (Tl) having a first main electrode coupled to the reference terminal (8), a second main electrode coupled to the input terminal (6) and a control electrode coupled to the first node (18); the third transconductance amplifier (32) comprises a second transistor (T3) having a first main electrode coupled to a second node (48), a second main electrode coupled to a supply voltage terminal (52) and a control electrode coupled to a bias voltage terminal (50), a third transistor (T4) having a first main electrode coupled to the reference terminal (8), a second main electrode coupled to the first node (18) and a control electrode coupled to the second node (48); and a fourth transistor (T2) having a first main electrode coupled to the reference terminal (8), a second main electrode coupled to the second node (48) and a control electrode coupled to the first node (18); the second transconductance amplifier (12) comprises a fifth transistor (T6) having a first main electrode coupled to the reference terminal (8), a second main electrode coupled to the output terminal (10) and a control electrode coupled to the first node (18); and the fourth transconductance amplifier (12) comprises a sixth transistor (T5) having a first main electrode coupled to the first node (18), a second main electrode coupled to a supply voltage terminal (52) and a control electrode coupled to the input terminal (6).

5. An electronic circuit as claimed in claim 3 or 4, further comprising a chain of two diode-connected transistors (T7, T8) series-connected between the bias voltage terminal (50) and the reference terminal (8) and a bias current source (54) coupled to the chain for providing a bias current through the chain.

6. An electronic circuit as claimed in claim 1, 2, 3, 4 or 5, further comprising a resistor (4) for coupling the input terminal (6) to a signal voltage source (2).