GB2319372A

GB2319372A - Bi-MOS multiplier

Info

Publication number: GB2319372A
Application number: GB9803110A
Authority: GB
Inventors: Katsuji Kimura
Original assignee: NEC Corp
Current assignee: NEC Corp
Priority date: 1993-11-10
Filing date: 1994-11-10
Publication date: 1998-05-20
Anticipated expiration: 2014-11-10
Also published as: GB9803110D0; GB2319372B

Abstract

A Bi-MOS multiplier is provided which can enable linearity of the output differential current of cross-coupled, source-coupled pairs of the multiplier to be improved to enable a wide input voltage range to be obtained.

Description

Bi-MOS MULTIPLIER BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a multiplier for multiplying two input signals, and more particularly, to a multiplier operable within wider input voltage ranges and to be realized on bipolar-MOS (Bi- MOS) integrated circuits.

2. Description of the Prior Art Fig. 1 shows a conventional OTA of this type that has the simplest configuration, which is disclosed in Ph.D. discertion, University of California, Berkeley, CA, 1985, entitled "High-frequency CMOS continuous time filters" written by H. Khorramabadi. This OTA is composed of first and second source-coupled pairs of MOS field-effect transistors (MOSFET) whose drains are cross-coupled, and is termed an MOSOTA.

As shown in Fig. 1, n-channel MOS transistors Mli and M12 with the same transconductance parameter p1 form a first balanced differential pair that is driven by a first constant current source (current: 1ssi) N-channel MOS transistors M13 and M14 with the same transconductance parameter p2 form a second balanced differential pair that is driven by a second constant current source (current: ISS2), where pi S p2.

Sources of the first and second transistors M11 and M12 are coupled together to be connected to the first current source. Sources of the third and fourth transistors M13 and M14 are coupled together to be connected to the second current source.

Gates of the transistors M11 and M13 are coupled together to be connected to one end 11 of an input end pair. Gates of the transistors M12 and M14 are coupled together to be connected to the other end 12 of the input end pair. An input voltage Vi is applied across the pair of input ends 11 and 12.

Drains of the transistors M11 and M14 are coupled together and drains of the transistors M12 and M13 are coupled together. A differential output current Al of the conventional MOSOTA is derived from the coupled drains of the transistors M11 and M14 and those of the transistors M12 and M13.

The differential output current Al is obtained by the following way: Here, drain currents of the transistors Mali, M12, M13 and M14 are defined as ID111 ID12, 1D13' ID14, respectively; then, the differential output current Al can be expressed as Al = (ID11 + ID14). - (ID12 + ID13).

The transconductance parameters pi and ss2 are defined as p1 = (C0x/2)(W1/Ll) and P2 = (Cox/2)(W2/L2) where is the effective surface mobility of a carrier, Cox is the gate oxide capacitance per unit area, W1 and L1 are a gate-width and a gate-length of the transistors M11 and M12 and W2 and L2 are a gate-width and a gate-length of the transistors M13 and M14, respectively. Also, COX is expressed as (#OX/tOX) where #OX and tox are the dielectric constant and the thickness of the gate oxide, respectively.

A differential output current #I11 (= ID11 - ID12) of the first balanced differential pair is expressed by the following equations (la) and (1b) as

Similarly, a differential output current #I12 (= 1D14 - 1D13) of the second balanced differential pair is expressed by the following equations (2a) and (2b) as

Here, assuming that (ISS1/ss1) > (ISS2/ss1) since generality is not lost, the differential output current Al of the conventional MOSOTA can be expressed by the following equation (3) as

where IV, (ISS2/ss2) 1/2 The equation (3) can be approximated by the following equation (4) that is obtained by using an approximation equation disclosed in ICICE Transactions on Electronics, Vol.

E76-C, No. 5, pp 720,May 1993, entitled "A Unified Analysis of Four-Quadrant Analog Multipliers Consisting of Emitter- and Source-Coupled Transistors Operable on Low Supply Voltage" written by the inventor, K. KIMURA.

To make the trans conductance of the MOSOTA linear in the equation (3), all of the quadratic and higher terms of the input voltage Vi need to be zero. This means that the cubic term of Vi, i.e., Vi3 in the equation (4) needs to be zero.

Therefore, the necessary condition for making the trans conductance of the conventional MOSOTA linear can be expressed by the following equation (5) as

The input-output and trans conductance characteristics of the conventional MOSOTA in Fig. 1, which are obtained under the above condition (5), are shown in Figs. 2 and 3, respectively.

The transconductance characteristic curves T1 to T12 shown in Fig. 3 are obtained under the conditions in the table 1 shown below where

Table 1

T1 0 0.49 T2 0.242 ISS1 0.495 ss1 T3 0.277 ISS1 0.566 p1 T4 0.318 ISS1 0.566 ss1 0.5625 T5 0.375 ISS1 0.667 ss1 T6 0.398 ISS1 0.707 ss1 T7 0.5916 0.377 ISS1 0.637 ss1 T8 0.410 ISS1 0.586 ss1 T9 0.471 ISS1 0.674 ss1 0.7 T10 0.512 ISS1 0.732 p1 T11 0.525 ISS1 0.751 ss1 T12 0.81 0.680 ISS1 0.839 ss1 From the characteristics shown in Figs. 2 and 3, it is seen that the trans conductance only changes within about 4 % over 70 % of the operable input voltage range or more, and as a result, the linearity of the transconductance characteristic is improved in a sufficient wide input voltage range without using a complex circuit.

By the way, to realize the conventional MOSOTA on an LSI, it is required that the transconductance parameter ratio (P2/P1) i.e., (W2/L2)/(W1/L1) has a specified value and that the driving current ratio (ISS2/ISS1) also has a specified value. Further, to make the ratios possibly exact, the values of the transconductance parameter ratio (P2/P1) and the driving current ratio (ISS2/ISS1) need to be either natural numbers or ratios of natural numbers, respectively.

Therefore, unit MOS transistors has to be employed in order to realize at least one of a desired value of the transconductance parameter ratio (p2/ssl) and a desired value of the driving current ratio (Iss2/Issl) which increases the number of the transistors incorporated and the chip occupation area of the conventional MOSOTA.

On the other hand, Fig. 4 shows a conventional Bi-MOS multiplier, which is composed of cross-coupled, emittercoupled pairs 50 of npn bipolar transistors Q11, Q12, Q13 and Q14 and a source-coupled pair 60 of MOS field-effect transistors M15 and M16. The cross-coupled, emitter-coupled pairs 50 are applied with a first input voltage Vl and the source coupled pair 60 is applied with a second input voltage v2.

The source-coupled pair 60 is driven by a constant current source (current: iso) In detail, the cross-coupled, emitter-coupled pair 50 is composed of a first emitter-coupled pair of npn transistors Q11 and Q12 whose collectors are coupled together and a second emitter-coupled pair of npn transistors Q13 and Q14 whose collectors are coupled together.

The coupled collectors of the transistors Q11 and Q13 are connected to one end 56 of an output end pair. The coupled collectors of the transistors Q12 and Q14 are connected to the other end 57 of the output end pair. A differential output current hIoz of the conventional Bi-MOS multiplier is derived from the pair of the output ends 56 and 57.

Bases of the transistors Q11 and Q14 are coupled together to be connected to one end 51 of a first input end pair.

Bases of the transistors Q12 and Q13 are coupled together to be connected to the other end 52 of the first input end pair.

The first input voltage V1 is applied across the first pair of the input ends 51 and 52.

The source-coupled pair 60 is composed of n-channel MOS transistors M15 and M16 whose sources are coupled together to be connected to the constant current source.

A drain of the transistor M15 is connected to the coupled emitters of the bipolar transistors Q11 and Q12. A drain of the transistor M16 is connected to the coupled emitters of the bipolar transistors Q13 and Q14. A differential output current hIlo is outputted from the drains of the MOS transistors M15 and M16 to drive the cross-coupled, emitter-coupled pairs 50.

A gate of the MOS transistor M15 is connected to one end 61 of a second input end pair. A base of the transistor M16 is connected to the other end 62 of the second input end pair. The second input voltage V2 is applied across the second pair of the input ends 61 and 62.

The differential output current AIoz of the conventional Bi-MOS multiplier is expressed by the following equations (6a) and (6b) as

In the equations (6a) and (6b), a is the dc common-base current gain factor of an npn bipolar transistor, p is the transconductance parameters of the MOS transistors M15 and M16, and VT is the thermal voltage that is expressed as VT = kT/q where k is Boltzmann's constant, T is absolute temperature in degrees Kelvin and q is the charge of an electron.

From the equations (6a) and (6b), it can be confirmed that in the cross-coupled, emitter-coupled pairs 50, the nonlinearity of the differential output current tIoz is -7.6 % when Vl = 2VT SO that the absolute value of the first input voltage V1 is limited to less than 2VT, i.e., 1V1I < 2Vr.

It can also be confirmed that in the source-coupled pair 60, the input voltage range for the second input voltage V2 is decided by a ratio of the driving current 1o and the transconductance parameter ss, i.e., (I0/ss), so that the nonlinearity of the differential output current hI1o is 7 % or less when the second input voltage V2 is less than 0.5.((2I0)/ss), i,e., |V2| < 0.5.[(2I0)/ss] As described above, with the conventional Bi-MOS multiplier in Fig. 4, as shown in the equations (6a) and (6b), the driving current 1o needs to be increased in order to widen the input voltage range for the second input voltage V2.

There are other related prior art as follows: The Japanese Non-Examined Patent Publication No. 60-146371 (August, 1985) discloses a CMOS analog multiplier with a wide dynamic range. The CMOS analog multiplier contains a first differential pair of first and second MOSFETs whose sources are coupled together and a second differential pair of third and fourth MOSFETs whose sources are coupled together.

The coupled sources of the first and second MOSFETs are connected to a first constant current sink and the coupled sources of the third and fourth MOSFETs are connected to a second constant current sink.

Gates of the first and second MOSFETs are coupled together to be connected to one end of a first input end pair. Gates of the third and fourth MOSFETs are coupled together to be connected to the other end of the first input end pair. A first input voltage to be multiplied is applied across the first input end pair.

Substrates of the first and second MOSFETs are coupled together to be connected to one end of a second input end pair. Substrates of the third and fourth MOSFETs are coupled together to be connected to the other end of the second input end pair. A second input voltage to be multiplied is applied across the second input end pair.

Drains of the first and third MOSFETs are coupled together to be connected to a first load. Drains of the second and fourth MOSFETs are coupled together to be connected to a second load. A first input voltage to be multiplied is applied across the first input end pair.

The Japanese Non-Examined Patent Publication No. 61-105912 (May, 1986) discloses a mixer circuit that can be easily formed on semiconductor integrated circuits and that can provide a sufficient conversion gain even while the input signal is small in amplitude.

The mixer circuit contains a double-balanced multiplier with two inputs and one output and a differential amplifier for amplifying two input signals and applying the input signals thus amplified to the multiplier differentially. The multiplier and the differential amplifier are composed of bipolar transistors, respectively.

The Japanese Non-Examined Patent Publication No. 3-4615 (January, 1991) discloses a multiplier in which an improved efficiency for taking out the frequency component of a clock signal can be obtained.

The multiplier contains first and second differential pairs of bipolar transistors whose respective emitters have resistors in order to widen the linear range of the inputoutput characteristics.

The Japanese Non-Examined Patent Publication No. 3-75977 (March, 1991) discloses a multiplier in which an output with a square-law characteristic can be obtained efficiently even if a difference between the dc biases to positive- and opposite-phase input signals.

The multiplier contains first and second differential amplifiers of bipolar transistors that are driven by the respective driving currents equal in value to each other provided at an input stage of the multiplier. First and second outputs are derived from first and second load resistances of the differential amplifiers. The first and second outputs are inputted into the multiplier through emitter followers, respectively.

An object of at least the preferred embodiments of the present invention is to provide a Bi-MOS multiplier in which a wider input voltage range can be obtained than the conventional one shown in Fig. 4 without increase in driving current.

According to the present invention, a Bi-MOS multiplier is provided, which contains a cross-coupled, emitter-coupled pairs applied with a first input voltage and a cross-coupled, source-coupled pairs applied with a second input voltage. The cross-coupled, emitter-coupled pairs are driven by a differential output current of the cross-coupled, source-coupled pairs.

The cross-coupled, emitter-coupled pairs are composed of a first differential pair of first and second bipolar transistors whose emitters are coupled together and a second differential pair of third and fourth bipolar transistors whose emitters are coupled together.

The collectors of the first and third bipolar transistors are coupled together and the collectors of the second and fourth bipolar transistors are coupled together. A differential output current of the multiplier is derived from the coupled collectors of the first and third transistors and the coupled collectors of the second and fourth transistors.

Bases of the first and fourth transistors are coupled together and bases of the second and third transistors are coupled together. The first input voltage is applied across the coupled bases of the first and fourth transistors and the coupled bases of the second and third transistors.

The cross-coupled, source-coupled pairs are composed of a third balanced differential pair of first and second MOS field-effect transistors whose sources are coupled together, and a fourth balanced differential pair of third and fourth MOS field-effect transistors whose sources are coupled together.

The first and second MOS transistors have the same transconductance parameter p11 and the third and fourth MOS transistors have the same transconductance parameter p12.

The coupled sources of the first and second MOS transistors are connected to a first constant current source whose constant current is 101, and the coupled sources of the third and fourth MOS transistors are connected to a second constant current source whose constant current is Iso2.

Drains of the first and fourth MOS transistors are coupled together to be connected to the coupled emitters of the first and second bipolar transistors. Drains of the second and third MOS transistors are coupled together to be connected to the coupled emitters of the third and fourth bipolar transistors. The differential output current of the cross-coupled, source-coupled pairs is outputted from the coupled drains of the first and fourth MOS transistors and the coupled drains of the second and third MOS transistors.

Gates of the first and third MOS transistors are coupled together and gates of the second and fourth transistors are coupled together. The second input voltage is applied across the coupled gates of the first and third MOS transistors and the coupled gates of the second and fourth transistors.

The currents 101 and 102 of the first and second constant current sources and the transconductance parameters ss11 and p12 of the first, second, third and fourth MOS transistors have such a relationship as

With the Bi-MOS multiplier of the invention, since the currents 101 and 102 of the first and second constant current sources and the trans conductance parameters P11 and p12 have the above relationship, linearity of the output differential current of the cross-coupled, source-coupled pairs is improved. As a result, a wider input voltage range can be obtained than the conventional one shown in Fig. 4 with no increase of the currents Iol and I02.

The Bi-MOS multiplier of the present invention is different from the prior-art CMOS analog multiplier disclosed in the Japanese Non-Examined Patent Publication No.

60-146371 because the prior-art CMOS analog multiplier is composed of only MOSFETs.

Also, the Bi-MOS multiplier. of the present invention is different from both of the prior-art mixer circuit disclosed in the Japanese Non-Examined Patent Publication No. 61-105912 and the prior-art multipliers disclosed in the Japanese Non-Examined Patent Publication Nos.

3-4615 and 3-75977 because the prior-art ones are composed of only bipolar transistors, respectively.

BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a circuit diagram showing a conventional MOSOTA.

Fig. 2 is a graph showing the input-output characteristic of the conventional MOSOTA shown in Fig. 1.

Fig. 3 is a graph showing the transconductance characteristic of the conventional MOSOTA shown in Fig. 1.

Fig. 4 is a circuit diagram showing a conventional Bi-MOS multiplier.

Fig. 5 is a circuit diagram of an MOSOTA.

Fig. 6 is a circuit diagram of a Bi-MOS multiplier according to a first embodiment of the invention.

Fig. 7 is a graph showing the input-output characteristic of the Bi-MOS multiplier shown in Fig. 6.

Fig. 8 is a circuit diagram of a Bi-MOS multiplier.

Fig. 9 is a circuit diagram of a Bi-MOS multiplier, according to a second embodiment of the invention.

Fig. 10 is a circuit diagram of another Bi-MOS multiplier.

Fig. 11 is a circuit diagram of a Bi-MOS multiplier, according to a third embodiment of the invention.

Fig. 12 is a circuit diagram of yet another Bi-MOS multiplier.

Fig. 13 is a circuit diagram of a Bi-MOS multiplier, according to a fourth embodiment of the invention.

Fig. 14 is a circuit diagram of another Bi-MOS multiplier.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below.

As shown in Fig. 5, a MOSOTA is composed of a first balanced differential pair of n-channel MOS field-effect transistors M1 and M2 whose sources are coupled together and a second balanced differential pair of n-channel MOS fieldeffect transistors M3 and M4 whose sources are coupled together.

These four transistors M1, M2, M3 and M4 have the same capability i.e., tr ansconductance parameter ss3.

The coupled sources of the first and second transistors M1 and M2 are connected to a first constant current source (current: 1553), so that the first balanced differential pair is driven by the first current source. Similarly, the coupled sources of the third and fourth transistors M3 and M4 are connected to a second constant current source (current: ISS4), so that the second balanced differential pair is driven by the second current source, where ISS3 * 1554.

The first and second current sources are arranged between the first and second differential pairs and the ground, respectively.

Drains of the transistors M1 and M4 are coupled together and drains of the transistors M2 and M3 are coupled together.

A differential output current #I1, of the MOSOTA is derived from the coupled drains of the transistors M1 and M4 and those of the transistors M2 and M3.

Gates of the transistors M1 and M3 are connected through a first resistor (resistance: R11) to each other. Gates of the transistors M2 and M4 are connected through a second resistor (resistance: R11) to each other. Gates of the transistors M3 and M4 are connected through a third resistor (resistance: R12) to each other.

The gate of the transistor M1 is connected to one end 1 of an input end pair and the gate of the transistor M2 is connected to the other end 2 of the input end pair. An input voltage V1 is supplied to the pair of the input ends 1 and 2 to be applied directly across the gates of the transistors M1 and M2.

A voltage produced by dividing the input voltage Vj, i.e., (V1/C1), is applied across the gates of the transistors M3 and M4, where C1 is a constant (C1 > 1).

The differential output current Al1 of the MOSOTA is obtained by the following way: Here, drain currents of the transistors M1, M2, M3 and M4 are defined as ID1B ID2 ID3 and ID4, respectively; then-, the differential output current #I1 can be expressed as #I1 = (1D1 + ID4) - (ID2 + ID3) The transconductance parameter ssl is defined as p3 = (Cox/2)(W3/L3) where W3 and L3 are a gate-width and a gatelength of the transistors M1, M2, M3 and M4, respectively.

A differential output current Al1 (= 1D1 - ID2) of the first balanced differential pair is expressed by the following equations (6a) and (6b) as

Similarly, a differential output current hI2 (= ID4 - ID3) of the second balanced differential pair is expressed by the following equations (7a) and (7b) as

The constant, i.e., divide ratio C1 has the following relationship with the resistances R11 and R12 as

Therefore, compared with the equations (2a) and (7a) under the condition of p, = ss2, it is seen that these equations (2a) and (7a) are in accordance with each other if the following relationship (9) is established.

If the divide ratio Ci, or the resistances R11 and R12 are decided so that the relationship (9) is established, the circuit configuration shown in Fig. 5 becomes equivalent to that shown in Fig. 1, which provides the same input-output characteristic shown in Fig. 2 and the same transconductance characteristic shown in Fig. 3 as those of the conventional one.

The first, second and third resistors may be made of patterned polysilicon films, respectively. In general, resistors made of the patterned polysilicon films do not increase the distortion of the input signal passing through the resistors if the resistors are produced through popular fabrication process steps. Additionally, such polysilicon resistors do not enlarge the chip areas of the transistors M1, M2, M3 and M4.

Also, almost voluntary values of the resistance R11 and R12 can be realized because the values will be decided by the minimum measurement or size of a mask used in their fabrication process step.

Further, since the constant Cl is expressed by a ratio of the resistances R11 and R12 as shown in the equation (8), obtainable value of Cl is small in fluctuation due to the fabrication processes.

As described above, the applied voltage to the input ends of the MOS transistors M3 and M4 changes according to the value of the divide ratio C1. As a result, a difference between the transconductance parameters of the first balanced differential pair and that of the second balanced differential pair is substantially generated, which is equivalent to the circuit configuration of the conventional one shown in Fig. 1.

Accordingly, an equivalent input-output characteristic and an equivalent transconductance characteristic to those of the conventional one in Fig. 1 can be obtained.

Because the MOS transistors M1, M2, M3 and M4 has the same transconductance parameter p3, the MOSOTA can be realized by common-sized transistors without increase in chip area.

Fig. 6 shows a Bi-MOS multiplier according to a first embodiment of the invention.

The Bi-MOS multiplier of the first embodiment contains cross-coupled, emitter-coupled pairs 10 applied with a first input voltage Vl and cross-coupled, source-coupled pairs 20 applied with a second input voltage V2. The cross-coupled, emitter-coupled pairs 10 are driven by a differential output current tIll of the cross-coupled, source-coupled pairs 20.

The cross-coupled, emitter-coupled pairs 10 are composed of a first differential pair of npn bipolar transistors Q1 and Q2 whose emitters are coupled together and a second differential pair of npn bipolar transistors Q3 and Q4 whose emitters are coupled together.

Collectors of the bipolar transistors Q1 and Q3 are coupled together to be connected to one end 16 of an output end pair. Also, collectors of the bipolar transistors Q2 and Q4 are coupled together to be connected to the other end 17 of the output end pair. A differential output current AIowll of the multiplier is derived from the pair of the output ends 16 and 17.

Bases of the transistors Q1 and Q4 are coupled together td be connected to one end 11 of a first input end pair.

Bases of the transistors Q2 and Q3 are coupled together to be connected to the other end 12 of the first input end pair.

The first input voltage vl is applied across the first input end pair 11 and 12.

The cross-coupled, emitter-coupled pairs 10 described above are substantially the same in configuration as the cross-coupled, emitter-coupled pairs 50 of the conventional Bi-MOS multiplier shown in Fig. 4.

The cross-coupled, source-coupled pairs 20 are composed of a third balanced differential pair of n-channel MOS fieldeffect transistors M5 and M6 whose sources are coupled together, and a fourth balanced differential pair of nchannel MOS field-effect transistors M7 and M8 whose sources are coupled together.

The MOS transistors M5 and M6 have the same transconductance parameter p11 and the MOS transistors M7 and M8 have the same transconductance parameter p12 where ss11 g p12.

The coupled sour

Drains of the MOS transistors M5 and M8 are coupled together to be connected to the coupled emitters of the bipolar transistors Q1 and Q2. Drains of the MOS transistors M6 and M7 are coupled together to be connected to the coupled emitters of the bipolar transistors Q3 and Q4.

The differential output current #I11 of the cross-coupled, source-coupled pairs 20 are outputted from the coupled drains of the MOS transistors M5 and M8 and the coupled drains of the MOS transistors M6 and M7.

Gates of the MOS transistors M5 and M7 are coupled together to be connected to one end 21 of a second input end pair. Gates of the transistors M6 and M8 are coupled together to be connected to the other end 22 of the second input end pair. The second input voltage V2 is applied across the second pair of the input ends 21 and 22.

The differential output current #I11 of the cross-coupled, source-coupled pairs 20 are, if the input voltage range of V2 is limited, expressed by the following equation (10) as

The equation (10) can be approximated by the following equation (11) as

The equation (11) provides a superior approximation of the equation (10) due to its little error.

To make the differential output current Al11 linear relative to the second input voltage V2, it is seen from the equation (11) that the following relationship (12) needs to be satisfied.

The MOS transistors MS and M6 with the same transconductance parameter ss11 have the same ratio (W/L)1 of their gate-widths W and gate-lengths L. Similarly, the MOS transistors M7 and M8 with the same transconductance parameter p12 have the same ratio (W/L)2 of their gate-widths W and gate-lengths L. Further, the transconductance parameter ratio (ss11/ss12) can be expressed using the ratios (W/L)1 and (W/L)2.

Therefore, the relationship (12) can be rewritten as

Accordingly, if the driving currents I01 and Ioz of the first and second constant current sources and the transconductance parameters ss11 and p12 of the MOS transistors M5, M6, M7 and M8 have the above relationship (12) or (13), the differential output current Al11 can be improved in linearity.

Fig. 7 shows the input-output characteristic of the Bi MOS multiplier of the first embodiment satisfying the above relationship (12) or (13). It is seen from Fig. 7 that the differential output current #I11 is linear within a relatively wide range of the second input voltage V2.

Since the cross-coupled, emitter-coupled pairs 10 are driven by the differential output current Al11 thus improved in linearity, a wider input voltage range can be obtained than the conventional one shown in Fig. 4 with no increase of the driving currents Io1 and I02.

In the first embodiment, the transconductance parameters p,, and p12r in other words, the ratios (W/L)1 and (W/L)2 of the gate-widths W and gate-lengths L, are different from each other. However, the transconductance parameters ss11 and ss12, or the ratios (W/L)l and (W/L)2 may be the same as each other.

Fig. 8 shows a Bi-MOS multiplier.

Similar to the multiplier shown in Fig. 6, the Bi-MOS multiplier shown in Fig. 8 contains cross-coupled, emittercoupled pairs 10 applied with a first input voltage vl and cross-coupled, source-coupled pairs 20a applied with a second input voltage V2. The cross-coupled, emitter-coupled pairs 10 are driven by a differential output current Al12 of the cross-coupled, source-coupled pairs 20a.

The cross-coupled, emitter-coupled pairs 10 are the same in configuration as those of Fig. 6, so that the description thereof is omitted here.

The cross-coupled, source-coupled pairs 20a are composed of a third balanced differential pair of n-channel MOS fieldeffect transistors M5' and M6' whose sources are coupled together, and a fourth balanced differential pair of nchannel MOS field-effect transistors M7' and M8' whose sources are coupled together.

The MOS transistors M5', M6', M7' and M8' have the same transconductance parameter P3 The coupled sources of the MOS transistors M5' and M6' are connected to a first constant current source whose constant current is ILl. The coupled sources of the MOS transistors M7' and M8' are connected to a second constant current source whose constant current is Il2. The first and second current sources are arranged between the third and fourth differential pairs and the ground, respectively.

Drains of the MOS transistors M5' and M8' are coupled together to be connected to the coupled emitters of the bipolar transistors Q1 and 02. Drains of the MOS transistors M6' and M7' are coupled together to be connected to the coupled emitters of the bipolar transistors Q3 and Q4.

The differential output current hIl2 of the cross-coupled, source-coupled pairs 20a is outputted from the coupled drains of the MOS transistors M5' and M8' and the coupled drains of the MOS transistors M6' and M7'.

Gates of the MOS transistors M5' and M7' are coupled together through a first resistor (resistance: R3) to each other. Gates of the MOS transistors M6' and M8' are coupled together through a second resistor (resistance: R3) to each other. Gates of the MOS transistors M7' and M8' are coupled together through a third resistor (resistance: R4) to each other where R3 # R4.

The gate of the MOS transistor MS' is connected to one end 21 of a second input end pair and the gate of the MOS transistor M6' is connected to the other end 22 of the second input end pair. A second input voltage V2 is supplied to the second pair of the input ends 21 and 22 to be applied directly across the gates of the MOS transistors M5' and M6'.

A voltage produced by dividing the second input voltage V2 i.e. (V2/C2) is applied across the gates of the MOS transistors M7' and M8', where C2 is a divide ratio (C2 > 1).

The divide ratio C2 can be expressed by the following equation (11) as R4 2 = (2R3 +R4) (11) The differential output current AIl2 of the cross-coupled, source-coupled pairs 20a is, if the input voltage range of V2 is limited, expressed by the following equation (12) as

The equation (12) can be approximated by the following equation (13) as

To make the differential output current #I12 linear relative to the second input voltage v2, it is seen from the equation (13) that the following relationship (14) needs to be satisfied.

Therefore, the multiplier shown in Fig. 8 has an equivalent circuit configuration to that of they first embodiment, if the divide ratio C2 satisfies the above relationship (14) and the transconductance parameter ratio (ss11/ss12) in the first embodiment shown in Fig. 6 have the following relationship (15) as

This means that the Bi-MOS multiplier shown in Fig. 8 embodiment has the same effect or advantage as that shown in Fig. 6.

Fig. 9 shows a Bi-MOS multiplier according to a second embodiment of the invent ion, which is substantially the same in configuration as that of the first embodiment in Fig. 6 other than that the conductivity types of the respective transistors used therein are opposite to each other.

In detail, the Bi-MOS multiplier of the second embodiment contains cross-coupled, emitter-coupled pairs 10A applied with a first input voltage vl and cross-coupled, source-coupled pairs 20A applied with a second input voltage V2. The crosscoupled, emitter-coupled pairs 10A are driven by a differential output current Al13 of the cross-coupled, source-coupled pairs 20A.

The cross-coupled, emitter-coupled pairs 10A are composed of a first differential pair of pnp bipolar transistors Qla and Q2a whose emitters are coupled together and a second differential pair of pnp bipolar transistors Q3a and Q4a whose emitters are coupled together.

Collectors of the bipolar transistors Qla and Q3a are coupled together to one end 16 of the output end pair. Also, collectors of the bipolar transistors Q2a and Q4a are coupled together to be connected to the other end 17 of the output end pair. A differential output current hIo=13 of the multiplier is derived from the output pair of the ends 16 and 17.

Bases of the transistors Qla and Q4a are coupled together to be connected to one end 11 of a first input end pair.

Bases of the transistors Q2a and Q3a are coupled together to be connected to the other end 12 of the first input end pair.

The first input voltage Vl is applied across the first input pair of the ends 11 and 12.

The cross-coupled, source-coupled pairs 20A are composed of a third balanced differential pair of p-channel MOS fieldeffect transistors M5a and M6a whose sources are coupled together, and a fourth balanced differential pair of pchannel MOS field-effect transistors M7a and M8a whose sources are coupled together.

The MOS transistors M5a and M6a have the same transconductance parameter p21 and the MOS transistors M7a and M8a have the same transconductance parameter p22, where ss21 $ p22.

The coupled sources of the MOS transistors M5a and M6a are connected to a first constant current source whose constant current is 121. The coupled sources of the MOS transistors M7a and M8a are connected to a second constant current source whose constant current is 122. The first and second current sources are arranged between the third and fourth differential pairs and a constant voltage source (voltage: V=).

Drains of the MOS transistors M5a and M8a are coupled together to be connected to the coupled emitters of the bipolar transistors Qla and Q2a. Drains of the MOS transistors M6a and M7a are coupled together to be connected to the coupled emitters of the bipolar transistors Q3a and Q4a.

Gates of the MOS transistors M5a and M7a are coupled together to be connected to one end 21 of a second input end pair. Gates of the transistors M6a and M8a are coupled together to be connected to the other end 22 of the second input end pair. The second input voltage V2 is applied across the second input end pair 21 and 22.

If the driving currents 121 and 122 of the first and second constant current sources and the transconductance parameters p21 and p22 of the MOS transistors M5a, M6a, M7a and M8a have such a~relationship as the above expression (12) or (13), the differential output current Al13 of the crosscoupled, source-coupled pairs 20A can be improved in linearity.

Therefore, similar to the first embodiment, a wider input voltage range of V2 can be obtained than the conven tional one shown in Fig. 4 with no increase of the driving currents 121 and 122.

Fig. 10 shows a Bi-MOS multiplier which is substantially the same in configuration as that shown in Fig. 8 other than that the conductivity types of the respective transistors used therein are opposite to each other.

In detail, similar to multiplier shown in Fig. 8, this Bi-MOS multiplier contains the cross-coupled, emitter-coupled pairs 10A applied with a first input voltage Vl, which are the same in configuration as those shown in Fig. 9, and cross-coupled, source-coupled pairs 20Aa applied with a second input voltage V2. The cross-coupled, emittercoup led pairs 10A are driven by a differential output current Al14 of the cross-coupled, source-coupled pairs 20Aa.

The cross-coupled, source-coupled pair 2OAa is composed of a third balanced differential pair of p-channel MOS fieldeffect transistors M5a' and M6a' whose sources are coupled together, and a fourth balanced differential pair of pchannel MOS field-effect transistors M7a' and M8a' whose sources are coupled together.

The MOS transistors M5a', M6a', M7a' and M8a' have the same transconductance parameter p23.

The coupled sources of the MOS transistors M5a' and M6a' are connected to a first constant current source whose constant current is 131. The coupled sources of the MOS transistors M7a' and M8a' are connected to a second constant current source whose constant current is I32. The first and second current sources are arranged between the third and fourth differential pairs and a constant voltage source (voltage: Vcc).

Drains of the MOS transistors M5a' and M8a' are coupled together to be connected to the coupled emitters of the bipolar transistors Qla and Q2a. Drains of the MOS transistors M6a' and M7a' are coupled together to be connected to the coupled emitters of the bipolar transistors Q3a and Q4a.

The differential output current Awl14 of the cross-coupled, source-coupled pairs 20Aa is outputted from the coupled drains of the MOS transistors M5a' and M8a' and the coupled drains of the MOS transistors M6a' and M7a'.

Gates of the MOS transistors M5a' and M7a' are coupled together through a first resistor (resistance: R1) to each other. Gates of the MOS transistors M6a' and M8a' are coupled together through a second resistor (resistance: R1) to each other. Gates of the MOS transistors M7a' and M8a' are coupled together through a third resistor (resistance: R2) to each other where R1 t R2.

The gate of the MOS transistor MSa' is connected to one end 21 of a second input end pair and the gate of the MOS transistor M6a' is connected to the other end 22 of the second input end pair. A second input voltage V2 is supplied to the second input end pair 21 and 22 to be applied directly across the gates of the MOS transistors M5a' and M6a'.

If the driving currents 131 and I32 of the first and second constant current sources and the trans conductance parameter p23 of the MOS transistors M5a' M6a', M7a' and M8a' have such a relationship as the above expression (12) or (13), the differential output current #I14 of the crosscoupled, source-coupled pairs 20Aa can be improved in linearity. Therefore, similar to the multiplier shown in Fig. 8, a wider input voltage range of V2 can be obtained than the conventional one shown in Fig. 4 with no increase of the driving currents I31 and 132.

Fig. 11 shows a Bi-MOS multiplier according to a third embodiment of the invention, which is equivalent to a combination of the cross-coupled, emitter-coupled pairs 10 shown in Fig. 6 and the cross-coupled, source-coupled pairs 20A shown in Fig. 9.

As shown in Fig. 11, a first constant current source (current: 141) is connected to the coupled sources of the MOS transistors M5a and M6a and a second constant current source (current: 142) is connected to the coupled sources of the MOS transistors M7a and M8a. The first current source is disposed between the third differential pair of the MOS transistors M5a and M6a and a constant voltage source (voltage: Vcc). The second constant current source is disposed between the fourth differential pair of the MOS transistors M7a and M8a and the constant voltage source.

Additionally, in the sixth embodiment, there is a third constant current source (current: I41 - I42) between the coupled emitters of the transistors Q1 and Q2 and the ground.

Also, there is a fourth constant current source (current: 141 - 142) between the coupled emitters of the transistors Q3 and Q4 and the ground.

A differential output current #I15 of the cross-coupled, source-coupled pairs 20A is derived from the coupled drains of the transistors M5a and M8a and those of the transistors M6a and M7a.

A differential output current hIozls of the multiplier is derived from the pair of the output ends 16 and 17.

Because the multiplier of the third embodiment is substantially the same in configuration as that of the first embodiment in Fig. 6, the same effect or advantage as that of the second embodiment can be obtained.

Further, there is an additional advantage that the multiplier of the third embodiment can operate a lower supply voltage than that of the first embodiment. However, there arises a disadvantage that the total current of the multiplier becomes about three times as much as that of the first embodiment.

Fig. 12 shows another Bi-MOS multiplier, which is equivalent to a combination of the cross-coupled, emitter-coupled pairs 10 shown in Fig. 6 and the cross-coupled, source-coupled pairs 20aA shown in Fig. 10.

As shown in Fig. 12, a first constant current source (current: its1) is connected to the coupled sources of the MOS transistors M5a' and M6a' and a second constant current source (current: 152) is connected to the coupled sources of the MOS transistors M7a' and M8a'. The first current source is disposed between the third differential pair of the MOS transistors M5a' and M6a' and a constant voltage source (voltage: Vow). The second constant current source is disposed between the fourth differential pair of the MOS transistors M7a' and M8a' and the constant voltage source.

Additionally, there is a third constant current source (current: I51 - I52) between the coupled emitters of the transistors Q1 and Q2 and the ground. Also, there is a fourth constant current source (current: I51 - I52) between the coupled emitters of the transistors Q3 and Q4 and the ground.

A differential output current hI,6 of the cross-coupled, source-coupled pairs 20Aa is derived from the coupled drains of the transistors M5a' and M8a' and those of the transistors M6a' and M7a'.

A differential output current hIoUrl6 of the multiplier is derived from the pair of the output ends 16 and 17.

Because of the multiplier shown in Fig. 12 is substantially the same in configuration as that shown in Fig. 8, the same effect or advantage as that of the multiplier shown in Fig. 8 can be obtained.

Further, there is an additional advantage that the multiplier can operate a lower supply voltage than that shown in Fig. 8. However, there arises a disadvantage that the total current of the multiplier becomes about three times as much as that shown in Fig. 8.

Fig. 13 shows a Bi-MOS multiplier according to a fourth 'embodiment of the invention, which is equivalent to a combination of the cross-coupled, emitter-coupled pairs 10A shown in Fig. 9 and the cross-coupled, source-coupled pairs 20 shown in Fig. 6.

As shown in Fig. 13, a first constant current source (current: Iso) is connected to the coupled sources of the MOS transistors MS and M6 and a second constant current source (current: 162) is connected to the coupled sources of the MOS transistors M7 and M8. The first current source is disposed between the third differential pair of the MOS transistors M5 and M6 and a constant voltage source (voltage: V). The second constant current source is disposed between the fourth differential pair of the MOS transistors M7 and M8 and the constant voltage source.

A differential output current #I17 of the cross-coupled, source-coupled pairs 20 is derived from the coupled drains of the transistors M5 and M8 and those of the transistors M6 and M7.

A differential output current AIozl7 of the multiplier is derived from the pair of the output ends 16 and 17.

Because the multiplier of the fourth embodiment is substantially the same in configuration as that of the first embodiment in Fig. 6, the same effect or advantage as that of the first embodiment can be obtained.

Fig. 14 shows another Bi-MOS multiplier, which is equivalent to a combination of the cross-coupled, emitter-coupled pairs 10A shown in Fig. 10 and the cross-coupled, source-coupled pairs 20a shown in Fig. 8.

As shown in Fig. 14, a first constant current source (current: 171) is connected to the coupled sources of the MOS transistors M5' and M6' and a second constant current source (current: 172) is connected to the coupled sources of the MOS transistors M7' and M8'. The first current source is disposed between the third differential pair of the MOS transistors M5' and M6' and a constant voltage source (voltage: V=). The second constant current source is disposed between the fourth differential pair of the MOS transistors M7' and M8' and the constant voltage source.

A differential output current #I17 of the cross-coupled, source-coupled pairs 20a is derived from the coupled drains of the transistors M5' and M8' and those of the transistors M6' and M7'.

A differential output current hIoUrl7 of the multiplier is derived from the pair of the output ends 16 and 17.

Because the multiplier shown in Fig. 14 is substantially the same in configuration as that shown in Fig. 8, the same effect or advantage as that shown in Fig. 8 can be obtained.

Whilst the preferred forms of the present invention have been described, it is to be understood that modifications will be apparent to those skilled in the art without departing from the spirit of the invention. The scope of the invention, therefore, is to be determined solely by the following claims.

Each feature disclosed in this specification (which term includes the claims) and/or shown in the drawings may be incorporated in the invention independently of other disclosed and/or illustrated features.

Claims

CLAIMS 1. A Bi-MOS multiplier for multiplying first and second input voltages, comprising: (a) cross-coupled, emitter-coupled pairs applied with said first input voltage; (b) cross-coupled, source-coupled pairs applied with said second input voltage; (c) said cross-coupled, emitter-coupled pairs being driven by a differential output current of said cross-coupled, source-coupled pairs; and (d) said cross-coupled, emitter-coupled pairs being composed of a first differential pair of first and second bipolar transistors whose emitters are coupled together and a second differential pair of third and fourth bipolar transistors whose emitters are coupled together; said collectors of said first and third bipolar transistors being coupled together and said collectors of said second and fourth bipolar transistors being coupled together, a differential output current being derivable from said coupled collectors of said first and third transistors and said coupled collectors of said second and fourth transistors; bases of said first and fourth transistors being coupled together and bases of said second and third transistors being coupled together, said first input voltage to be applied across said coupled bases of said first and fourth transistors and said coupled bases of said second and third transistors; (e) said cross-coupled, source-coupled pairs being composed of a third balanced differential pair of first and second MOS field-effect transistors whose sources are coupled together, and a fourth balanced differential pair of third and fourth MOS field-effect transistors whose sources are coupled together; said first and second MOS transistors having the same transconductance parameter ssll and said third and fourth MOS transistors having the same transconductance parameter ssl2; said coupled sources of said first and second MOS transistors being connected to a first constant current source whose current is t and said coupled sources of said third and fourth MOS transistors being connected to a second constant current source whose constant current is I02; drains of said first and fourth MOS transistors being coupled together to be connected to said coupled emitters of said first and second bipolar transistors, and drains of said second and third MOS transistors being coupled together to be connected to said coupled emitters of said third and fourth bipolar transistors; said differential output current of said cross-coupled, source-coupled pairs to be output from said coupled drains of said first and fourth MOS transistors and said coupled drains of said second and third MOS transistors; gates of said first and third MOS transistors being coupled together and gates of said second and fourth transistors being coupled together, said second input voltage to be applied across said coupled gates of said first and third MOS transistors and said coupled gates of said second and fourth transistors; and said currents I01 and b, of said first and second constant current sources and said transconductance parameters ss11 and ssl2 of said first, second, third and fourth MOS transistors have such a relationship as
2. A multiplier as claimed in Claim 1, further comprising a third constant current source whose current is (I01 - 102) connected to said coupled emitters of said first and second bipolar transistors; and a fourth constant current source whose current is (Iol - Io2) connected to said coupled emitters of said third and fourth bipolar transistors.
3. A Bi-MOS multiplier substantially as herein described with reference to any of Figures 6, 9, 11 and 13 of the accompanying drawings.