JPH0583139A - Switched capacitor/continuous time buffer circuit for digital/analog converter - Google Patents

Abstract

PURPOSE: To convert a sample data analog signal by a switched capacitor into a continuance analog signal by a D/A converter substantially without increasing distortion. CONSTITUTION: A digital input is converted into an analog value in a sample data area and then inputted to a quaternary switched capacitor/continuance buffer. This buffer has an amplifier 160 which has a switched capacitor 166 connected to its input and a feedback capacitor 174 connected across the switched capacitor 166. The input side of this capacitor 166 is switched between an input sample data voltage and the amplifier output. Consequently, those capacitors 166 and 174 reach substantially equal voltages at low frequencies, so a request for electric charges to the amplifier 160 is not present at charging time. Consequently, this amplifier 166 inputs no distortion when an output signal sampled by the switched capacitor 166 is converted into the continuance input signal of an active filter.

Description

Detailed Description of the Invention

[0001]

FIELD OF THE INVENTION This invention relates generally to switched capacitor filters, and more particularly to a digital-to-analog converter using a switched capacitor filter and an active filter to filter its output.

[0002]

BACKGROUND OF THE INVENTION Conventional digital-to-analog converters are provided with a multi-stage low-pass filter for filtering out unwanted noise after the first conversion stage. There is a type that uses a delta-sigma digital modulator as one type of digital-analog converter.
The delta-sigma modulator receives a digital input and converts it into a 1-bit digital output. Normally, this output is applied to an active RC low pass filter after passing through a 1-bit DAC (digital-analog converter). This active RC
A low pass filter is a series resistor and various active components connected to achieve the desired filtering. This type of filter has the disadvantage of being sensitive to variations in these components. Therefore, it takes a lot of time and effort for fine adjustment.

One type of filter is an optimized switched capacitor filter. However, this type of filter does not have its noise characteristic optimized to its limit because the switched capacitor filter of the DAC using the delta-sigma modulator is composed of the bi-quad switched capacitor filters connected in cascade. This leads to the problem that a large number of biquad stages introduces a considerable amount of unwanted noise components into the overall filtering action.

A problem with any type of conventional digital-to-analog converter that uses a switched capacitor filter is the need to convert the output of the switched capacitor filter from the sample data domain to the continuous time domain. In the past, the active filter was used to directly convert the sampled data output from the switched capacitor into a continuous time output. However, this transformation usually introduces large distortions in the signal.

The present invention provides a digital-to-analog converter. The digital-to-analog converter has an input that receives an n-bit digital input signal. The converter further includes circuitry for converting the n-bit digital input signal to a sample data analog signal, and a sample data-continuous time buffer circuit receiving the sample data analog signal without substantially increasing distortion of the signal. Convert to continuous time analog signal.

As a second feature of the present invention, the conversion circuit has a switched capacitor filter for filtering the sample data analog signal. further,
It has a delta-sigma digital modulator, which is n
The bit digital signal is converted into an m-bit digital signal smaller than n. Further, an m-bit digital-analog converter is provided, which converts the m-bit digital signal into a sample data analog signal and inputs it to the switched capacitor filter. In the preferred embodiment, this m is equal to 1. The output of the sample data / continuous time buffer circuit is filtered by a low pass filter.

As a further feature of the present invention, the sample data / continuous time buffer circuit includes an amplifier having a finite bandwidth and a feedback capacitor connected between the input and the output. There is also an input capacitor controlled by the switching circuit. The switching circuit is operable to switch one side of the input capacitor between the sample data input analog signal and the output of the amplifier and the other side of the capacitor between the reference voltage and the input of the amplifier. This switching circuit
It further comprises a first switching circuit section connecting one side of the input capacitor to the sample data analog signal at the first time point and to the output of the amplifier at the second time point. The switching circuit is also provided with a second circuit part, which connects the other side of the input capacitor to the reference voltage at the first time point and to the input of the amplifier at the second time point.

The present invention will be described in detail below with reference to the accompanying drawings.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT FIG. 1 is an overall block diagram of a digital-to-analog converter using a low noise / low distortion switched capacitor / continuous time filter. An n-bit digital input is applied to the input of interpolation filter 11 and the output of that filter is applied to the input of digital modulator 10. The digital modulator is operable to convert an n-bit input into an m-bit digital output smaller than n. Digital Modulator 10 of the Preferred Embodiment
Consists of an oversampling / noise shaping circuit using a delta-sigma digital modulator that converts an n-bit digital word into an m-bit digital output. In the preferred embodiment m is equal to 1. However, it is also possible to use a multi-bit output digital modulator.

Although the preferred embodiment uses a delta-sigma modulator, any type of 1-bit digital modulator or its equivalent could be used to convert an n-bit digital word into a 1-bit digital signal stream. .. Delta-sigma modulators are used because of their good low-level characteristics and their good differential non-linearity. The general operation of digital modulator 10 is known in the art.

The 1-bit digital signal output from the digital modulator 10 is sent to the fourth-order switched capacitor low-pass filter 12. This filter 12 is a Butt that integrates a 1-bit DAC into its input, as will be described later.
It is an erworth type filter. The output of this filter 12 is input to a switched capacitor / continuous time buffer circuit 14 which converts the output of the capacitor switching into a continuous time format with relatively low distortion. This signal is then input to a high impedance, low distortion unity gain buffer 16 whose output is input to an active RC low pass filter 18. The output of the active RC low pass filter 18 forms the low impedance analog output of the overall digital to analog converter of FIG.

FIG. 2 shows a schematic block diagram of the switched capacitor low pass filter 12. 1-bit digital output of delta-sigma modulator 10 is 1-bit DA
Input to C20. The output of the 1-bit DAC 20 is input to the positive input of the addition point 22, and its output is connected to the integrator 24 of the first stage. The negative input of summing point 22 is connected to output node 26 of lowpass filter 12 as a whole.

The output of the integrator 24 of the first stage is input to the positive input of the addition point 28, and its negative input is connected to the node 26. The output of the addition point 28 is the second integrator 3
Input to 0. The output of the second-stage integrator 30 is the addition point 3
Input to 2 positive inputs. The negative input of summing point 32 is connected to output node 26. The output of the addition point 32 is 3
It is connected to the input of the integrator 34 of the stage. The output of the integrator 34 of the third stage is input to the positive input of the addition point 36.
The negative input of summing point 36 is connected to node 26.
The output of the addition point 36 is input to the fourth-stage integrator 38, whose output is connected to the node 26.

In the preferred embodiment, addition points 22, 2
8 and 32 have two positive inputs and the third stage integrator 34
Has an inverted configuration. Therefore, the fourth stage integrator 38
Gives the inverted output from. This is equivalent to the structure of FIG.

The filter of FIG. 2 constitutes a 4th order Butterworth type filter using a switched capacitor topology. By utilizing the switched capacitor topology, the fabrication of the filter can be easily done at the current state of the art. Relatively low noise operation is obtained by using a high order filter with feedback from the output stage to the input stage. The transfer function of the filter of FIG. 2 is as follows.

[0016]

[Equation 1] FIG. 3 shows the topology of the switched capacitor.
3 is a more detailed block diagram of the filter of FIG. The 1-bit digital signal input is input to switched capacitor / DAC block 40, and the output of that block is input to summing node 42. The addition node 42 corresponds to the addition point 22. The output of node 42 is input to the negative input of amplifier 44, whose positive input is grounded. A capacitor 46 is connected between the negative input and output of amplifier 44. The output of amplifier 44 is input to switched capacitor block 48, whose output is connected to summing node 50. The addition node 50 is the addition point 28
Is input to the negative input of the amplifier 52. The positive input of amplifier 52 is grounded. A capacitor 54 is connected between the negative input and output of amplifier 52. The amplifier 52 and the capacitor 54 correspond to the second stage integrator 30.

The output of amplifier 52 is input to a switched capacitor block 56 whose output is connected to summing node 58. The addition node 58 corresponds to the addition point 32. Summing node 58 is input to the negative input of amplifier 60, whose positive input is grounded.
A capacitor 62 is connected between the negative input and output of amplifier 60. The amplifier 60 and the capacitor 62 correspond to the third stage integrator 34. The output of the amplifier 60 is input to the switched capacitor block 64, whose output is connected to the summing node 66 corresponding to the summing point 36. Summing node 66 is input to the negative input of amplifier 68, whose positive input is grounded. A capacitor 70 is connected between the negative input and output of amplifier 68. The amplifier 68 and the capacitor 70 correspond to the integrator 38 of the fourth stage. The output of amplifier 68 is output node 26
It is connected to the.

Output node 26 is fed back to summing nodes 42, 50, 58, 66 via switched capacitor blocks 72, 74, 76, 78, respectively.
By feeding back the output of the final stage integrator from node 26 to the input of amplifier 44, input stage 24 can influence the output noise of final stage integrator 38. Therefore, 1 corresponding to the amplifier 44 and the capacitor 46
The gain of the second stage integrator 24 reduces the noise associated with the inputs of all subsequent stages, including the output stage integrator 38. The result is a very low noise switched capacitor filter.

This filter operates at a sampling rate of 3.072 MHz and has a bandwidth of -3 dBz of 25.
The transfer function of the filter, which is Khz and is normalized to a bandwidth of 1 / 2π-3 dB, is as follows.

[0020]

[Equation 2] Switched capacitor block 40, 48, 56, 6
Capacitor 46 used for 4, 72, 74, 76, 78,
The capacitance values of 54, 62, and 70 are the same as those of the capacitors C1-C1.
For 2 is given as follows. C1 corresponds to the switched capacitor block 42 and C2 is the block 70.
, C3 to the capacitor 46, C4 to the block 48,
C5 is in block 74, C6 is in capacitor 54, C7
Corresponds to block 56, C8 to block 76, C9 to capacitor 62, C10 to block 64, C11 to block 78, and C12 to capacitor 70. The values of the capacitors C1-C12 are as follows. FIG. 4 shows a switched capacitor / DAC block 40,
A detailed logic diagram of the switched capacitor block 72 and amplifier 44 is shown. The reference voltage is connected to node 80,
This node 80 also has a node 84 via a switch 82.
It is connected to the. Node 84 is connected to one side of capacitor 86 and is also grounded via switch 88. The other side of capacitor 86 is connected to node 90, which is connected to summing node 42 via switch 92. Node 90 is also grounded via switch 94 and further grounded via switch 96.

Like the switch 94, the switch 82 is controlled by the clock φ1. Switch 88 is clock φ2
Controlled by. Switch 96 is controlled by the AND logic function of clock .phi.2 and the reciprocal of the single bit digital input "1BIT". The switch 92 is controlled by the AND function of the clock φ2 and the input 1BIT.

The reference voltage at node 80 is also a switch 98.
Is input to the node 100 via. Node 100 is connected to one side of capacitor 102, and switch 1
It is grounded via 04. The other side of capacitor 102 is connected to node 106. Node 10
6 is connected to the addition node 42 via the switch 108, and is also grounded via the switches 110 and 112.

The switches 104 and 110 have a clock φ1.
And the switch 98 is controlled by the clock φ2. The switch 112 has a clock φ2 and an input signal 1
It is controlled by the AND function of BIT. Switch 10
8 is the AND of the clock φ2 and the reciprocal of the input signal 1BIT
Controlled by function. Due to this switching, the path associated with capacitor 86 is the non-inverting path and the path for capacitor 102 is the inverting path.

The output of the node 26 is input to the node 118 via the switch 116. Node 118 is connected to one side of capacitor 120, and also switch 1
It is grounded via 22. The other side of capacitor 120 is connected to node 124. Node 12
4 is connected to summing node 42 via switch 126 and is also grounded via switch 128.
The switches 116, 122, 126, 128 and the capacitor 120 form the switched capacitor block 72. Since switches 116 and 128 are clocked by clock φ1 and switches 122 and 126 are clocked by clock φ2, a non-inverting topology is obtained as described above. In addition, the elements forming the switched capacitor block 72 are similar to the elements forming the switched capacitor blocks 48, 56, 64, 74, 76, 78, but the values of the capacitors are different, and the blocks 64, 78 are It has an inverted configuration. This is a conventional switched capacitor configuration.

FIG. 5 is a timing diagram showing the operation of the clocks φ1 and φ2 and the single bit input.
When clock .phi.1 is high, switches 82 and 94 are in the closed position, thus charging the plate of capacitor 86 connected to node 84 to the reference voltage. Clock φ
When 2 is high, node 84 is at ground potential,
Node 90 is connected to summing node 42 because switch 92 is closed only when the single bit digital input is a logic one. If not, switch 92 remains in the open position and switch 96 grounds node 90 causing capacitor 86 to discharge. Because switches 108 and 112 operate in the opposite manner, capacitor 1 is used to measure the voltage on summing point 42 when 1BIT is in the opposite logic state, a logic one. Of course, at the same time that charge is transferred from capacitor 86 or 102 to summing point 42, charge is also transferred from capacitor 120 to summing point 42. This is feedback, which causes the first-stage integrator to reduce noise in the output stage. This gives the topology of the entire 4th order Butterworth type filter using a 1-bit DAC integrated into the input of the first stage filter.

FIG. 6 is a block diagram of a digital / analog converter for explaining the operation of the switched capacitor / continuous time buffer circuit 14. Generally, the delta / sigma digital modulator 10 is incorporated into a digital / sample analog signal converter 130. The output of this block 130 is the analog value of the digital input signal in the sample data area. This is input to the switched capacitor / continuous time buffer 14, which
4 allows conversion to a continuous-time analog signal with virtually no increase in distortion. The unity gain amplifier 16 enables high impedance operation of the buffer 14. The output of amplifier 16 is input to an active low pass filter 18 capable of driving a relatively low impedance output on the order of 600 ohms.

FIG. 7 is a schematic diagram showing the operation of the buffer 14. Each data sample in the sample data area, that is, a sample, is represented by discrete steps. A sine wave was used for purposes of illustration. The only important data point in the sample time domain is at the end of the sample data value, which is the point at which the data is valid. It is necessary to convert this data into a smooth analog waveform that represents the output of buffer 14. However, this conversion must be performed relatively undistorted, and such conversion is difficult when the sampled data is generated with some slew as a result of the switched-capacitor filter not having ideal characteristics. .. This type of signal is usually obtained. Moreover, the use of conventional buffer circuits introduces more slews in the sample data domain, resulting in increased distortion on conversion to the continuous time domain.

FIG. 8 is a schematic diagram of a sample data waveform. The sample data is represented by a plurality of discrete steps 132. At the end of each step 132 is valid analog data for that value. These points must be connected to each other without distortion.

FIG. 9 shows a prior art non-inverting first order switched capacitor low pass filter which is insensitive to parasitic capacitance. The input of amplifier 134 is switched. The sampled input voltage of the switched capacitor is input to one side of the switch 136 while the other side is the capacitor 138.
Connected to one side. One side of capacitor 138 is also grounded via switch 140. The other side of the capacitor 138 is the switch 142.
Connected to one side of the switch and the other side of the switch to the negative input of amplifier 134. The other side of capacitor 138 is also grounded via switch 144. Switches 136 and 144 are clocks φ1
The switches 140 and 142 are controlled by the clock φ.
Controlled by 2.

The feedback circuit is provided with a feedback capacitor 146 connected between the negative input and output of the amplifier 134. The negative input of amplifier 134 is also connected to one side of switch 148, the other side of which is connected to one side of capacitor 150. One side of capacitor 150 is also grounded via switch 152. Capacitor 1
The other side of 50 is grounded through switch 154 and is also connected to one side of switch 156.
The other side of switch 156 is connected to the output of amplifier 134. Switches 152 and 154 are clock φ
1 and the switches 148 and 156 are controlled by the clock φ2.

Assuming that the filter configuration of FIG. 9 is configured to have unity low frequency gain (ie, the value of the input sampling capacitor 138 equals the switched feedback capacitor 150), the z-domain transfer function. Is as follows.

[0032]

[Equation 3] After the filter reaches steady state at the DC input, its output settles at the DC input value just before sampling occurs. Output voltage equals input voltage (equal to DC voltage) with equal value switched capacitors
In this case, this sampling does not affect the charge on the feedback capacitor 146 in the unswitched state. However, this amplifier 134 must generate an output carrying the charge required by the change in voltage polarity on the series capacitance of the capacitor 138 and the capacitor 150 whose value is substantially equal to this capacitor. This charge requirement may drive amplifier 134 to temporarily exhibit non-linear settling behavior, which is observed as distortion in subsequent continuous time blocks.

FIG. 10 is a schematic diagram of a switched capacitor / continuous time buffer circuit according to the present invention. Amplifier 1
The positive input of 60 is grounded and its negative input is connected to input node 162. One side of switch 164 receives the sampled input voltage by the switched capacitor, while the other side is connected to one side of capacitor 166. One side of capacitor 166 is also connected to the output of amplifier 160 via switch 168. The other side of capacitor 166 is grounded through switch 170 and also switch 172.
Is connected to the input node 162 via. Feedback capacitor 174 is connected to input node 162 and amplifier 1
The output of 60 is connected.

In operation, the buffer circuit of FIG. 10 eliminates the potential distortion generating mechanism described above with respect to FIG. 9 by reducing the charge requirements on the amplifier 160 during low frequency steady state operation. Since the single input capacitor 166 is switched directly between the input and the output, it will be on any capacitor when the output voltage equals the input voltage and the input voltage equals a relatively low frequency voltage such as direct current. There is no net voltage change. Therefore, there is no charge requirement on the output 160 of the amplifier. In this way the strain is relatively low. Since a high impedance load in the form of the unity gain buffer 16 is driven, the effect is further improved.

FIG. 11 is a schematic diagram of a prior art unity gain buffer circuit. Two differential input transistors 17
0, 172 are provided, and their sources are connected to the common node 174. This node 174 is connected to the supply voltage V _SSA via the source-drain path of the N-channel transistor 176. Since this gate is connected to the bias voltage, transistor 1
76 acts as a current source. The drain of transistor 170 is connected to one side of the source-drain path of P-channel transistor 178, and the other side is connected to the positive supply voltage V _DDA on node 180. The gate of transistor 178 is connected to its drain. Similarly, the drain of transistor 172 is connected to one side of the source-drain path of transistor 182 and the other side is connected to node 180. The gate of the transistor 182 is the transistor 17
8 gates connected. The gate of transistor 170 constitutes the negative input and the gate of transistor 172 constitutes the positive input. The drain of the transistor 172 becomes the output of the differential amplifier connected to the gate of the output drive transistor 184 which is a P-channel transistor, and one side of the source-drain path of this transistor 184 is connected to the node 180 and the other side. Are connected to output node 186. Output node 186 is connected to the gate of transistor 170 which is the input of this amplifier. The source-drain path of the transistor 188 serving as a current source is connected between the node 186 and the voltage V _SSA . The gate of transistor 188 is connected to the bias voltage.

In the differential input transistor pair 170, 172 of the prior art amplifier shown in FIG. 11, the large common mode signal creates two distortion generating mechanisms. The first mechanism is that the drain-source voltage (V _DS ) of the differential transistor pair 170 and 172 greatly changes depending on the signal. I _D vs. V _DS of these transistors,
That is, if there is a mismatch in the output conductance characteristics, it is necessary to introduce the difference in the gate-source voltage V _GS of these transistors to maintain the drain current (I _D ) at an equal value. This difference in V _GS appears as distortion in the input / output characteristics of the buffer. A second present in prior art amplifier / buffer circuits when the common mode signal is large.
The distortion generation mechanism is due to a large change in the drain-source voltage of the current source transistor 176 depending on the signal. Since the output impedance of the transistor 176 is finite, the bias current of the differential pair changes with the input signal due to the change of V _DS of the transistor 176. If there is a mismatch in the I _D vs. V _GS , or transconductance characteristics of the differential transistor pair, then the gate-source voltage difference of these devices must be reintroduced to maintain equal current balance. It is important to point out that the open-loop amplifier gain does not mitigate these effects at all during closed-loop operation, because both of these distortion generating mechanisms occur in the differential input stage of the amplifier.

FIG. 12 is a simplified schematic diagram of the unity gain buffer circuit of the DAC. Two differential input transistors 190, 200 are provided, both of which are N-channel devices and their sources are both connected to a common node 202. The gate of transistor 190 constitutes the positive input of the buffer and the gate of transistor 200 constitutes the negative input. Current source 204 is connected to the drain of transistor 190 to provide current I '. Similarly, current source 206 is connected to the drain of transistor 200 to provide equal current I '. A current source 208 is connected between the common node 202 and the supply voltage V _SSA to provide a 2I current sink. Another current source, 210, is connected directly to node 202 to provide a current Δ
Give I. The current source 210 is a transistor 190, 20
0 respective drains and respective current sources 204, 2
It is controlled by two control boxes 212 which exist between 0 and 06 to sense the current passing therethrough. Control box 2
Reference numeral 12 controls the current source 210 to adjust the value of ΔI so that the current flowing through the transistors 190 and 200 is maintained at a constant value I _D irrespective of the current variation of the current source 208. Therefore, the differential input transistor pair 190, 200
Is a constant I even when the input common mode signal is large.
_{Operates in D} mode. Two transistors 190, 20
The current sources 204, 206 can be activated to force a drain current equal to 0, while the current source 210 can be activated to absorb the current difference at the common node 202.

FIG. 13 is a more detailed circuit diagram of the unity gain buffer circuit of the DAC. The common node 202 is connected to one side of the source-drain path of transistor 214 and the other side is connected to V _SSA . The transistor 214 corresponds to the current source 208. The gate of this transistor 214 is connected to the bias voltage. The drain of transistor 190 is connected to node 218 through the source-drain path of cascode N-channel transistor 216. Transistor 21
The gate of 6 is connected to the dynamic bias node 220. Node 218 is connected to one side of the source-drain path of P-channel transistor 222 and the other side is connected to supply node 224 which is connected to V _DDA . The gate of transistor 222 is connected to the bias voltage and acts as current source 204. Similarly, the drain of transistor 200 is connected to node 228 via the source-drain path of N-channel cascode transistor 226. The gate of transistor 226 is connected to node 220. Node 228 is connected to one side of the source-drain path of P-channel transistor 230 and the other side is connected to node 224.
It is connected to the. The gate of transistor 230 is connected to the bias voltage and acts as current source 206.

One side of the source-drain path of the P-channel feedback transistor 232 is connected to the node 22.
4 and the other side is connected to node 220. The gate of the transistor 232 is connected to the node 218. Similarly, one side of the source-drain path of P-channel feedback transistor 234 is connected to node 224 and the other side is connected to node 22.
It is connected to 0. The gate of the transistor 234 is connected to the node 228. The drain and gate of the N-channel transistor 236 are connected to the node 220 and the source thereof is connected to the node 202.
The current ΔI is applied to 2 and the transistors 216 and 22
6 is given a dynamic gate bias.

Node 228 is connected to the gate of P-channel transistor 237 and its source-drain path is connected between node 224 and output node 238. Similarly, node 238 is connected to one side of the source-drain path of N-channel transistor 240 and the other side to V _SSA . The gate of the transistor 240 is connected to the gate of the second N-channel transistor 242.
Has its gate connected to the drain and its source is V _SSA
Connected to act as a current mirror. Transistor 24
The gate and drain of 2 are P-channel transistors 24
4 source-drain path connected to one side and the other side connected to node 224. The gate of the transistor 244 is connected to the node 218. Transistors 237, 240, 242, 244 act as a second amplification stage to form a general turnaround output stage. Although transistors 240 and 242 are nothing more than current mirrors, transistors 237 and 244 form the actual output stage and this structure is a two stage amplifier. Output node 2
38 is connected to the gate of the transistor 190 which is the input of this amplifier.

In operation, the ΔI current is controlled by the transistors 232 and 234 which perform feedback operation. Transistors 216 and 226 are cascode devices in which transistor 236 is a bias device. Transistors 216 and 226 keep the V _DS of the differential pair transistor at a constant value even when the common mode signal at the input is large. The shunt feedback transistors 232, 234 are operable to absorb the differential current at the common node 202. Transistor 214
When the bias current of the transistor changes due to a finite output impedance, the gate-source voltage of the feedback transistors 232 and 234 changes to adapt to this current change, but the change of V _GS of the transistors 232 and 234 is (gm / go). It is a small value divided by >> 1 (gm =
Transconductance of shunt feedback transistors 232 and 234; go = current source transistor 2
14 output conductance). Therefore, transistor 1
The current source transistors that actually determine the 90 and 200 input bias currents have a V that is substantially less than the input common signal.
Experience changes in _DS .

FIG. 14 shows a start-up circuit of the buffer circuit of FIG. The symbol B is added to the node 218 in FIG.
Node 220 is labeled A and node 228 is labeled C. The P-channel transistor 246 is a source-
The drain path is connected between nodes 224 and 248. The gate of transistor 246 is connected to node C at node 220. The source-drain path of P-channel transistor 250 is connected between nodes 224 and 248, and its gate is connected to node B at node 218. The source-drain path of the P-channel transistor 252 is connected between the node 224 and the connection point A of the node 228, and the gate thereof is connected to the node 248. Transistor 254
_Has a source-drain path connected between node 248 and V _SSA and its gate connected to the bias current. The circuit of FIG. 14 is operable to provide a start-up current to node A upon startup.

FIG. 15 is a schematic diagram of the output stage of an operational amplifier used in a conventional active filter of a DAC. Source of N-channel source follower transistor 258-
The drain path is connected between the voltage V _DDA and the node 260. Node 260 is connected to a low voltage source having current source 262. Node 260 forms an output node, which is grounded through resistor 264, which is a load resistor. Current flowing through the resistor 264 is _{I L,} the current flowing through the transistor 258 drain current I
It is the sum of _D and the load current I _L.

In the prior art circuit of FIG. 15, the current of the source follower transistor 258 changes due to the load current I _L depending on the signal. As a result, the V _{GS of} this device
Changes the square law, and this becomes the distortion of the V _out / V _in transfer characteristic. Furthermore, the standard source follower is A
Operating in the class mode, this means that the direct current source _ID must be as large as the intended maximum load current.

FIG. 16 is a simplified circuit diagram of the output stage of the active RC low pass filter 18 of the DAC. The source-drain path of source follower N-channel transistor 266 is connected between nodes 268 and 260, and its source is connected to node 260. The gate of transistor 266 receives the input voltage to this output stage. The constant current source 270 has a node 260 and a voltage V _SSA.
And a constant current source 272 is connected between node 268 and positive voltage V _DDA . Current source 27
The current flowing through 0, 272 is equal to the drain current I _D flowing through the transistor 266, which remains constant regardless of the load current I _L. Variable current source 274 is V
_A variable current source 276 is connected between _DDA and node 260, and a variable current source 276 is connected between node 260 and voltage V _SSA . Current source 274 is operable to provide current to load resistor 264 to increase the voltage change at node 260, and current source 276 sinks current from load resistor 264 to provide a low voltage or negative signal change. Is operable. Current source 274, 276
Is controlled by a feedback circuit 278 operable to generate class AB feedback controlled by the voltage at node 268. Therefore, the load current is
74, 276, the current sources 274, 276 provide and sink a load current I _L greater than the operating bias current.

In operation, the source follower transistor 266 is operated in constant current mode. The drain current is made independent of the load by the current source 272. Since the drain current of the N-channel transistor 266 is constant, its gate-source voltage is independent of the load, and an undistorted voltage is sent to the load resistor 264. Current I _L required by load resistor 264
Are supplied by current sources 274, 276. The AB feedback circuit 278 provides Class AB operation, which is controlled by the voltage at node 268.

FIG. 17 is a more detailed circuit diagram of the output stage of FIG. Current source 272 is comprised of a P-channel transistor 280 whose source-drain path is connected between voltage _VDDA and node 268. Similarly, the current source 270 is composed of an N-channel transistor 282 whose source-drain path is connected between the node 260 and the voltage V _SSA . The gates of transistors 280 and 282 are connected so that a different bias voltage is obtained.

The current source 274 is constituted by a P-channel transistor 284 having a source-drain path connected between the voltage V _DDA and the node 260 and a gate connected to the node 268. Current source 276 is comprised of an N-channel transistor 286 whose source-drain path is connected between node 260 and voltage V _SSA . The transistor 286 is controlled by the current mirror operation.

The feedback circuit comprises a P-channel transistor 288 whose source-drain path is connected between the emitter of the NPN bipolar transistor 290 and the drain of the N-channel transistor 292. The gate of transistor 292 is connected with its drain to the gate of transistor 286 and its source is connected to V _SSA to be the other side of the current mirror. The gate of transistor 288 is connected to the external AB bias signal. The base of the transistor 290 is connected to the node 268, and its collector is V
_It is connected to the _DDA .

In operation, class AB operation is provided by transistor 288 and bipolar transistor 290, which provides low impedance control of the source of transistor 288. Since this bipolar transistor has a large transconductance and therefore has no problem of back gate, it is used instead of the N-channel transistor. Transistor 28
The bias currents of 4 and 286 are controlled by the AB bias signal which provides the gate bias to transistor 288. The signal operation is performed by feedback controlling the gate potential of the transistor 284 with the voltage of the node 268. As the gate potential of the transistor 284 decreases, the current flowing through the transistor 284 increases and supplies the current to the load resistor 264. At the same time, the V _{GS of} transistor 288 is reduced, which reduces the current in transistors 286 and 292 that make up the mirror. As the gate potential of the transistor 284 of the node 268 increases, the current of the transistor 284 decreases and the transistor 28 forming a mirror with the transistor 288.
The current of 6, 292 increases to sink the load current.
Note that transistors 284 and 286 are capable of providing and sinking a load current I _L that is greater than their operating bias currents, and thus note that their operation is class AB.

FIG. 18 is a schematic diagram of one embodiment of a circuit for generating an AB bias signal. P-channel transistor 294 has its source-drain path connected between node V _DDA and one side of current source 296. The other side of current source 296 is connected to node V _SSA ,
A current I _B flows through this current source 296. The bipolar transistor 298 has a collector connected to V _DDA and an emitter connected to the source of the P-channel transistor 300. The base of the transistor 298 is connected to the gate and drain of the transistor 294. The gate and drain of transistor 300 are connected to one side of current source 302 and the other side thereof is connected to node V _SSA . Current source 302 provides current I _B. The circuit of FIG. 18 provides the signal path of transistors 284 and 286 with a dark current proportional to current I _B according to the following relational expression:

[0052]

[Equation 4] FIG. 19 shows another embodiment of the bias circuit for applying the AB bias. The P-channel transistor 304 has its source-drain path connected between the node V _DDA and the node 306. Node 306 is connected to node V _SSA via the source-drain path of N-channel transistor 308. The gate of transistor 308 is connected to the gate and drain of N-channel transistor 310 to form a current mirror, the source of which is connected to node V _SSA . Transistor 31
The drain of 0 is connected to the drain of P-channel transistor 312, its source is the emitter of bipolar transistor 314 and its base is transistor 304.
Connected to the gate. The collector of transistor 314 is connected to V _DDA . Transistor 312
Is connected to the output of the operational amplifier 316.
This amplifier 316 has its negative input connected to node 306 and its positive input grounded. The P-channel transistor 318 has a source-drain path connected to the node 320 and V
It is connected to _DDA . The gate of transistor 318 is connected to the output of operational amplifier 322, its positive input is connected to node 320, and its negative input is grounded. The node 320 is a current source 32 through which the current I _B flows.
4 to the voltage V _SSA . In this circuit, all of these ratio set bias currents are set so that the corresponding devices in the signal path operate at the same V _GS and V _DS .

In summary, the switched capacitor / continuous time buffer circuit of the present invention receives the switched capacitor filter output in the sample data domain and converts it into the continuous time domain substantially without distortion. This buffer circuit uses a filter having at least one feedback capacitor. The input of the switched capacitor is switched such that there is substantially no charge on the amplifier of the filter when switched in low frequency operation. This is accomplished by switching the input capacitor directly between the input and output of the filter's amplifier so that there is no net voltage change on any capacitor in the system.

[Brief description of drawings]

FIG. 1 is an overall block diagram of a low noise / low distortion switched capacitor / continuous time filter.

FIG. 2 is a logic block diagram of a switched capacitor filter.

FIG. 3 is a more detailed block diagram of the switched capacitor filter of FIG.

FIG. 4 is a detailed logic diagram of the input stage of the switched capacitor filter of FIG.

FIG. 5 is a timing diagram of a 1-bit DAC integrated in the input stage of a switched capacitor filter.

FIG. 6 is a block diagram of a switched capacitor / continuous time buffer.

FIG. 7 is a schematic diagram illustrating a conversion operation.

8 shows a smoothing operation obtained by the buffer circuit of FIG.

FIG. 9 shows a prior art buffer that receives sampled data at the input from the output of a switched capacitor filter and converts it into a continuous time analog input to the active filter.

FIG. 10 is a diagram of a switched capacitor / DAC
FIG. 6 is a logic diagram of a continuous time buffer circuit.

FIG. 11 is a schematic diagram of a prior art unity gain buffer circuit.

FIG. 12 is a simplified schematic diagram of a unity gain buffer circuit.

FIG. 13 is a detailed circuit diagram of a unity gain buffer circuit.

FIG. 14 is a schematic diagram of a startup circuit of a unity gain buffer circuit.

FIG. 15 is a schematic diagram of a prior art output stage of a DAC.

FIG. 16 is a simplified schematic diagram of an output stage of a DAC.

FIG. 17 is a detailed circuit diagram of the output stage of the DAC.

FIG. 18 is a schematic diagram showing one embodiment of a bias circuit for generating an AB bias in the output stage.

FIG. 19 shows another example of generating an AB bias.

[Explanation of symbols]

 10 Interpolation Filter 11 Delta-Sigma Digital Modulator 12 Switched Capacitor / Low Pass Filter 14 Switched Capacitor / Continuous Time Buffer Circuit 16 Unit Gain Buffer Circuit 18 Active Low Pass Filter 20 DAC 130 Digital-Sample Analog Signal Converter 278 Feedback Circuit

Claims (25)

[Claims] 1. A digital-analog converter for converting an input digital signal into a continuous-time analog output signal, an input for receiving an n-bit digital input signal, and a circuit for converting the n-bit digital input signal into a sample data analog signal. A digital-analog converter comprising a sample data / continuous time buffer circuit for converting a sample data analog signal into a continuous time analog signal without substantially increasing the distortion of the sample data analog signal. 2. The converter according to claim 1, wherein the conversion circuit has a switched capacitor filter for filtering a sample data analog signal. 3. A delta-sigma digital modulator in which the conversion circuit converts an n-bit digital input signal into an m-bit digital signal of a number smaller than n, and an m-bit input m-bit digital signal to a switched capacitor filter. The converter according to claim 2, further comprising an m-bit digital-analog converter for converting the sample data into an analog signal. 4. Converter according to claim 1, characterized in that m is equal to 1. 5. The converter according to claim 1, further comprising a continuous-time low-pass filter for receiving and filtering the output of the sample data / continuous-time buffer circuit. 6. The sample data / continuous time buffer circuit comprises an amplifier having a finite bandwidth, a feedback capacitor connected between the input and output of the amplifier, a sampling capacitor, and one side of the sampling capacitor. 2. A converter as claimed in claim 1, comprising a switching circuit for switching between the sample data input analog signal and the output of the amplifier and for switching the other side of the sampling capacitor between the reference voltage and the input of the amplifier. .. 7. A first switching circuit portion, wherein the switching circuit connects one side of the sampling capacitor to a sample data analog signal at a first time point and to an output of an amplifier at a second time point, and 7. A converter as claimed in claim 6, characterized in that it comprises a second switching circuit part connecting the other side to the reference voltage at the first time point and to the input of the amplifier at the second time point. 8. A digital-to-analog converter for converting an input digital signal into a continuous time analog output signal, wherein an input for receiving an n-bit digital input signal and an n-bit digital input signal for an m-bit digital signal smaller than n. Delta-sigma digital modulator for conversion, m-bit digital-analog converter for converting m-bit digital signal to sample data analog signal, switched capacitor filter for filtering sample data analog signal, and distortion of sample data analog signal 1. A digital-analog converter comprising a sample data / continuous time buffer circuit for converting a sample data analog signal into a continuous time analog signal without increasing the number of samples. 9. Converter according to claim 8, characterized in that m is equal to 1. 10. The converter according to claim 8, further comprising a low pass filter for receiving an input of the sample data / continuous time buffer circuit and filtering the input. 11. The buffer circuit comprises: an amplifier having a finite bandwidth; a feedback capacitor connected between an input and an output of the amplifier; a sampling capacitor; and a sampling data analog signal on one side of the sampling capacitor. 9. A converter as claimed in claim 8 comprising a switching circuit switching between the output of the amplifier and the other side of the sampling capacitor between the reference voltage and the input of the amplifier. 12. A first switching circuit section, wherein said switching circuit connects one side of a sampling capacitor to an input sample data analog signal at a first time point and to an output of an amplifier at a second time point, and a sampling capacitor. 12. The converter of claim 11 comprising a second switching circuit section connecting the other side of the to the reference voltage at the first time and to the input of the amplifier at the second time. 13. A circuit for converting a sample data analog signal into a continuous time analog output signal, the input receiving the sample data analog signal and the sample data analog signal without substantially increasing the distortion of the sample data analog signal. A circuit comprising a sampled data / continuous time buffer circuit for converting the signal into a continuous time analog signal. 14. The sample data / continuous time buffer circuit comprises: an amplifier having a finite bandwidth, a feedback capacitor connected between the input and the output of the amplifier, a sampling capacitor, and one side of the sampling capacitor. 15. A switching circuit for switching between the sample data input analog signal and the output of the amplifier and the other side of the sampling capacitor between the reference voltage and the input of the amplifier. converter. 15. A first switching circuit portion, wherein the switching circuit connects one side of the sampling capacitor to a sample data analog signal at a first time point and to an output of an amplifier at a second time point, and 15. A converter as claimed in claim 14, characterized in that it comprises a second switching circuit part connecting the other side to the reference voltage at the first time point and to the input of the amplifier at the second time point. 16. A method for converting a digital signal into a continuous time analog output signal, the method comprising: receiving an n-bit digital input signal, converting the n-bit digital input signal into a sample data analog signal, and receiving a sample data analog input signal. And a step of converting the sampled data analog signal into a continuous time signal without substantially increasing the distortion of the sampled data analog signal. 17. The method of claim 16, wherein converting the n-bit digital input signal to a sample data analog signal comprises filtering the sample data analog signal with a switched capacitor filter. 18. The step of converting an n-bit digital input signal to a sample data analog signal inputs the n-bit digital input signal to a delta-sigma digital modulator and the n-bit digital input signal is delta-sigma digital modulated. 18. The method according to claim 17, further comprising a step of converting the signal into an m-bit digital signal smaller than n, converting the m-bit digital signal into a sample data analog signal, inputting the sampled data analog signal into a switched capacitor filter, and filtering the sampled analog signal. the method of. 19. The method of claim 18, wherein m is equal to 1. 20. The method of claim 16 including the step of converting a continuous time signal from a sample data analog signal and then filtering with a low pass filter. 21. The step of converting a sample data analog signal to a continuous time data signal provides an amplifier having a finite bandwidth, connecting a feedback capacitor between the input and the output of the amplifier, and And switching one side of the sampling capacitor between the sample data analog signal and the output of the amplifier and switching the other side of the sampling capacitor between the reference voltage and the input of the amplifier. The method according to claim 16. 22. The switching step connects one side of the sampling capacitor to a sample data analog signal at a first time point and connects the one side of the sampling capacitor to an output of an amplifier at a second time point, 22. Connecting the other side of the sampling capacitor to a reference voltage at a first time point and connecting the other side of the sampling capacitor to an input of an amplifier at a second time point.
The method described in. 23. A method of converting a sample data analog signal into a continuous time analog output signal, the method comprising receiving a sample data analog signal and converting the sample data analog signal into a continuous time analog signal without substantially increasing distortion. A method comprising the steps of: 24. The converting step provides an amplifier having a finite bandwidth, connecting a feedback capacitor between an input and an output of the amplifier, providing a sampling capacitor, and sampling one side of the sampling capacitor. 24. The method of claim 23, comprising the step of switching between the data analog signal and the output of the amplifier, and switching the other side of the sampling capacitor between the reference voltage and the input of the amplifier. 25. The switching step connects the one side of the sampling capacitor to a sample data input analog signal at a first time point and to the output of an amplifier at a second time point, and the other side of the sampling capacitor to a second side. 25. The method of claim 24, comprising connecting to a reference voltage at one time and to the input of the amplifier at a second time.