Comparator for a pipelined analog-to-digital converter and related signal sampling method

1. Field of the Invention

The present invention relates to a comparator for a pipelined analog-to-digital converter (ADC) and related signal sampling method, and more particularly, to a comparator that performs a sample-and-hold function as a front-end sample-and-hold amplifier in the pipelined ADC and related signal sampling method.

2. Description of the Prior Art

An analog-to-digital converter (ADC) converts an analog signal into a digital signal, and is frequently used in industrial measurement, communication system, and audio/video data processing. ADCs are implemented in the several types including flash ADC, successive approximation ADC, sigma-delta ADC, and pipelined ADC, which are recognized by different sampling rate and resolution. A pipelined ADC can provide both high sampling rate and high resolution, and is widely used in communication system.

Please refer to FIG. 1 , which is a block diagram of a 10-bit pipelined ADC 10 according to the prior art. The pipelined ADC 10 includes a sample-and-hold amplifier 100 , subsequent stages 102 _ 1 - 102 _ 8 following the sample-and-hold amplifier (SHA) 100 , a flash ADC 104 , and an error correction circuit 106 . The subsequent stages 102 _ 1 to 102 _ 8 are similar, and only the subsequent stage 102 _ 1 is illustrated as follows. The subsequent stage 102 _ 1 includes a sub-ADC 10 , a digital-to-analog converter (DAC) 112 , a sample-and-hold unit 114 , a subtractor 116 , and an amplifier 118 , where the DAC 112 , the sample-and-hold unit 114 , the subtractor 116 , and the amplifier 118 consist of a multiplying DAC.

Note that, the front-end sample-and-hold amplifier 100 is used to convert input analog signals into DC-like signals, such that quantization output of the sub-ADC 110 is not influenced easily by noise. However, the sample-and-hold amplifier 100 also brings distortion to the input analog signal of the sub-ADC 110 , and costs large power consumption in the pipelined ADC 10 . If the sample-and-hold amplifier 100 is not used, when the input analog signal is higher than tens of MHz, sampling points of the sub-ADC 110 may be different from that of the rear-stage multiplying DAC, which causes aperture error, and resolution of the pipelined ADC 10 is therefore reduced.

There are some conventional methods that omit the front-end sample-and-hold amplifier and also reserve the ample-and-hold effect, such as “A 55-mW, 10-bit, 40-Msample/s Nyquist-Rate CMOS ADC” by Iuri Mehr and Larry Singer, published in IEEE Journal of Solid-State Circuits, vol. 35, no. 3., which implements functions of the front-end sample-and-hold amplifier into the first subsequent stage of a pipelined ADC. Please refer to FIG. 2 , which is a schematic diagram of the first subsequent stage 20 of the pipelined ADC disclosed in the mentioned paper. The first subsequent stage 20 includes a sub-ADC 200 and a multiplying DAC 210 . The sub-ADC 200 includes two similar comparators 201 , 202 , and a logic circuit 204 , where detail of the comparator 202 is illustrated in FIG. 3 . In FIG. 2 , the pipelined ADC 210 is illustrated as a single-ended architecture for a simplified presentation, and in fact, the pipelined ADC 210 is a fully differential architecture.

As shown in FIG. 3 , the comparator 202 includes a preamplifier 206 , a latch circuit 208 , switches S 1 A, S 1 B, S 2 A, S 2 B, S 3 A, S 3 B, S 4 A, S 4 B, SZ 1 , SZ 2 , and capacitors C 1 -C 4 , wherein these switches and capacitors forms a switched capacitor circuit, as a sampling circuit, and reference voltages V TH1P , V TH1N , and a common mode voltage V CM are used in the switched capacitor circuit. The preamplifier 206 consists of transistors M 1 -M 6 . The multiplying DAC 210 includes a DAC 212 , an amplifier 214 , switches SM 1 -SM 5 , and capacitors CM 1 and CM 2 . For a pipelined ADC of a 1.5-bit/stage architecture, the multiplying DAC 210 requires a positive reference voltage V REFP , a negative reference voltage V REFN and the common mode voltage V CM . The reference voltage V TH1P that the switched capacitor circuit uses is equal to ¼ V REFP , and the reference voltage V TH1N is equal to ¼V REFN . Through the switched capacitor circuit, two threshold voltages, +¼ (V REFP −V REFN ) and −¼ (V REFP −V REFN ), are generated for the comparison of the comparator 202 . As shown in FIG. 3 , non-overlapping clock signals Φ 1 and Φ 2 are used as sampling clocks for two adjacent subsequent stages. Falling edges of clock signals Φ 1d and Φ 2d are later than falling edges of the clock signals Φ 1 and Φ 2 , respectively. A clock signal Φ 2c is a delay clock of the clock signal Φ 2d .

When in a sampling phase, the clock signal Φ 1 is at a high voltage level, and the switches S 1 A, S 2 A, S 3 A, and S 4 A are turned on. In this situation, the reference voltages V TH1P and V TH1N are respectively sampled to the capacitors C 1 -C 4 , differential input voltages V INP and V INN are sampled to the capacitors C 2 and C 3 , and the switches SZ 1 and SZ 2 are turned on in order to cancel a DC offset voltage across differential input terminals of the preamplifier 206 . In the multiplying DAC 210 , an input voltage V IN is sampled to the capacitors CM 1 and CM 2 when the clock signal Φ 1d is at a high voltage level.

When in a holding phase, the clock signal Φ 2d is at a high voltage level, and the switches S 1 B, S 2 B, S 3 B, and S 4 B are turned on. In this situation, a voltage (V INP −V INN ) is compared to the threshold voltages +¼ (V REFP −V REFN ) and −¼ (V REFP −V REFN ), and the voltage across the differential input terminals of the preamplifier 206 indicates a comparison result. The preamplifier 206 amplifies the voltage across the differential input terminals, and the latch circuit 208 latches differential output voltages of the preamplifier 206 at the rising edge of the clock signal Φ 2c . Next, the logic circuit 204 generates a 2-bit digital signal as 00, 10, or 11, which indicates different comparison result, according to output voltages latched by the latch circuits of the comparators 200 and 202 . In the multiplying DAC 210 , the DAC 212 outputs the reference voltages V REFP , V REFN , or V CM to the capacitor CM 2 according to the 2-bit digital signal outputted from the logic circuit 204 . Briefly, the multiplying DAC 210 amplifies the input voltage V IN and subtracts the outputted reference voltage (which is V REFP , V REFN , or V CM ) from the input voltage V IN , to generate a residue voltage V OUT that is outputted to a next subsequent stage.

From the above, the time when the preamplifier 206 performs amplification is between the rising edge of the clock signal Φ 2d and the rising edge of the clock signal Φ 2c , shown as T d in FIG. 3 . In order to prevent the aperture error, a time constant of the switched capacitor circuit in the sub-ADC 200 has to be equal to a time constant of the switched capacitor circuit in the multiplying DAC 210 . In other words, resistors and capacitors used in switched capacitor circuits in both sides should fulfill the following equation:

R M ⁢ C M = C 1 G m + 1 / R C ≈ C 1 G m , ( 1 )

where R M indicates the equivalent resistance of the switch SM 1 or SM 2 ; C M indicates the equivalent capacitance of the capacitor CM 1 or CM 2 ; G m is the transconductance of the transistor in the preamplifier 206 ; R C is the equivalent resistance of the switch in the switched capacitor circuit of the sub-ADC 200 . In high-speed applications, the above resistance R C or R M is quite small, so that the transconductance G m should be large enough to make both sides of the equation 1 to be equal, i.e., the time constants are matched. Note that, a large transconductance G m causes rise of power consumption of the comparator 202 . Also, it is difficult to obtain the matched time constants through adjusting the transconductance of a transistor because the transconductance is not easy to be controlled.

It is therefore a primary objective of the present invention to provide a comparator for a pipelined ADC and related signal sampling method.

The present invention discloses a comparator for a pipelined ADC that includes a sampling circuit coupled to a plurality of differential input voltages and a plurality of differential reference voltages, for sampling the plurality of differential input voltages according to a first clock signal and sampling the plurality of differential reference voltages according to a second clock signal, a preamplifier coupled to the sampling circuit comprising a positive input terminal, a negative input terminal, a positive output terminal, and a negative output terminal, for amplifying a voltage across the positive input terminal and the negative input terminal for generating a plurality of differential output voltages, and a latch circuit coupled to the preamplifier for latching the plurality of differential output voltages.

The present invention further discloses a signal sampling method for a pipelined ADC, wherein a comparator of the first subsequent stage of the pipelined ADC includes a sampling circuit including a plurality of input voltage sampling switches, a plurality of reference voltage sampling switches, and a plurality of capacitors. The signal sampling method includes controlling the plurality of input voltage sampling switches according to a first clock signal to sample a plurality of differential input voltages on the plurality of capacitors, and controlling the plurality of reference voltage sampling switches according to a second clock signal to sample a plurality of differential reference voltages on the plurality of capacitors, wherein the first clock signal and the second signal are non-overlapping signals.

These and other objectives of the present invention will no doubt become obvious to those of ordinary skill in the art after reading the following detailed description of the preferred embodiment that is illustrated in the various figures and drawings.

FIG. 1 is a block diagram of a 10-bit pipelined ADC according to the prior art.

FIG. 2 is a schematic diagram of the first subsequent stage of a pipelined ADC according to the prior art.

FIG. 3 is a schematic diagram of a comparator in FIG. 2 .

FIG. 4 is a schematic diagram of a comparator according to an embodiment of the present invention.

FIG. 5 is a schematic diagram of a subsequent stage of a pipelined ADC according to an embodiment of the present invention.

FIG. 6 is a schematic diagram of a comparator according to an embodiment of the present invention.

FIG. 7 and FIG. 8 are schematic diagrams of the second subsequent stage and the third subsequent stage behind the first subsequent stage shown in FIG. 5 .

FIG. 9 is a flowchart of a process according to an embodiment of the present invention.

Please refer to FIG. 4 , which is a schematic diagram of a comparator 400 according to an embodiment of the present invention. The comparator 400 performs functions as a sample-and-hold amplifier, and is used in a 1.5-bit sub-ADC. When the first subsequent stage of a pipelined ADC applies the comparator 400 , the pipelined ADC does not require using a front-end sample-and-hold amplifier, i.e., cost of the pipelined ADC can be reduced.

The comparator 400 comprises a preamplifier 402 , a latch circuit 404 , switches S 1 A, S 1 B, S 2 A, S 2 B, S 3 A, S 3 B, S 4 A, S 4 B, SZ 1 , SZ 2 , and S RST , capacitors C 1 -C 4 , and capacitors Cs 1 -Cs 4 . Compared to the conventional comparator 202 in FIG. 3 , the capacitors Cs 1 -Cs 4 and the switch S RST are added. The capacitors Cs 1 -Cs 4 is sampling capacitors, where the capacitance of Cs 1 -Cs 4 can be small and occupy small area. The switches S 1 A, S 1 B, S 2 A, S 2 B, S 3 A, S 3 B, S 4 A, S 4 B, the capacitors C 1 -C 4 , and Cs 1 -Cs 4 forms a switched capacitor circuit, also called a sampling circuit. A positive input voltage V INP and a negative input voltage V INN are differential input voltages of the comparator 400 , and other reference voltages coupled to the switched capacitor circuit are generated by a reference voltage generator of the pipelined ADC. Each switch in the switched capacitor circuit is utilized for controlling a signal connection between a reference voltage (or an input voltage) and one of the sampling capacitor Cs 1 -Cs 4 according to different clock signal, in order to sample the reference voltage (or the input voltage) on the sampling capacitor.

Components in the switched capacitor circuit are described as follows. The switch S 1 A is coupled to a positive reference voltage V REFP ; the switch S 1 B is coupled to the positive input voltage V INP ; one terminal of the capacitor C 1 is coupled to the switches S 1 A and S 1 B, and another terminal of the capacitor C 1 is coupled to an positive input terminal of the preamplifier 402 ; one terminal of the capacitor Cs 1 is coupled to the switches S 1 A, S 1 B, and the capacitor C 1 , and another terminal of the capacitor Cs 1 is coupled to a ground. The switch S 2 A is coupled to a negative reference voltage V REFN ; the switch S 2 B is coupled to the positive input voltage V INP ; one terminal of the capacitor C 2 is coupled to the switches S 2 A and S 2 B, and another terminal of the capacitor C 2 is coupled to the positive input terminal of the preamplifier 402 ; one terminal of the capacitor Cs 2 is coupled to the switches S 2 A, S 2 B, and the capacitor C 2 , and another terminal of the capacitor Cs 2 is coupled to the ground. The switch S 3 A is coupled to the positive reference voltage V REFP ; the switch S 3 B is coupled to the negative input voltage V INN ; one terminal of the capacitor C 3 is coupled to the switches S 3 A and S 3 B, and another terminal of the capacitor C 3 is coupled to an negative input terminal of the preamplifier 402 ; one terminal of the capacitor Cs 3 is coupled to the switches S 3 A, S 3 B, and the capacitor C 3 , and another terminal of the capacitor Cs 3 is coupled to the ground. The switch S 4 A is coupled to the negative reference voltage V REFN ; the switch S 4 B is coupled to the negative input voltage V INN ; one terminal of the capacitor C 4 is coupled to the switches S 4 A and S 4 B, and another terminal of the capacitor C 4 is coupled to the negative input terminal of the preamplifier 402 ; one terminal of the capacitor Cs 4 is coupled to the switches S 4 A, S 4 B, and the capacitor C 4 , and another terminal of the capacitor Cs 4 is coupled to the ground. As shown in FIG. 4 , the clock signals for controlling these switches are denoted beside the switches, which are described in detail later.

Note that, reference voltages used in the switched capacitor circuit of the comparator 400 are the positive reference voltage V REFP and the negative reference voltage V REFN . Threshold voltages used for comparison are +¼ (V REFP −V REFN ) and −¼ (V REFP −V REFN ), which are designed according to an ratio related to the capacitance of the capacitor C 1 -C 4 . According to the present invention, when the capacitor C 1 and the capacitor C 4 have the same capacitance, the capacitor C 2 and the capacitor C 3 have the same capacitance, and C 1 :C 2 =5:3, the threshold voltages +¼ (V REFP −V REFN ) and −¼ (V REFP −V REFN ) are therefore generated. Compared with the comparator 202 in FIG. 3 , the comparator 400 does not require using additional reference voltages as V TH1P and V TH1N and only uses the reference voltages V REFP and V REFN , to generate these threshold voltages. Since the reference voltage generator of the pipelined ADC does not need to generate additional reference voltages, power consumption of the pipelined ADC is reduced.

Please refer to FIG. 4 again for the detail of the preamplifier 402 . The preamplifier 402 comprises n-type transistors MN 1 -MN 4 and p-type transistors MP 1 -MP 4 , which are utilized for amplifying the voltage across the positive input terminal and the negative input terminal of the preamplifier 402 , for generating differential output voltages. The transistors MN 1 -MN 4 and MP 1 -MP 4 are metal-oxide-semiconductor field effect transistors (MOSFET), for example. Capacitors Cp 1 and Cp 2 shown in FIG. 4 are parasitic capacitors on differential input terminals of the preamplifier 402 . The transistors MN 1 and MN 2 consist of a current mirror. The drain of the transistor MN 1 is coupled to the gate of itself and a current source I IN , and the source of the transistor MN 1 is coupled to the ground. The gate of the transistor MN 2 is coupled to the gate of the transistor MN 1 , the source of the transistor MN 2 is coupled to the ground, and the drain of the transistor MN 2 is coupled to the source of the transistor MN 3 and the source of the transistor MN 4 .

The gate of the transistor MN 3 is the positive input terminal of the preamplifier 402 , and the gate of the transistor MN 4 is the negative input terminal of the preamplifier 402 . The sources of the transistors MP 1 -MP 4 are all coupled to a voltage source V CC . The gate of the transistor MP 1 is coupled to the drain of itself, the drain of the transistor MP 3 , and the drain of the transistor MN 3 , which is the negative output terminal of the preamplifier 402 . The gate of the transistor MP 2 is coupled to the drain of itself, the drain of the transistor MP 4 , and the drain of the transistor MN 4 , which is also the positive output terminal of the preamplifier 402 . In addition, the gate of the transistor MP 3 is coupled to the drain of the transistor MP 4 , and the gate of the transistor MP 4 is coupled to the drain of the transistor MP 3 , which forms an active load that makes gain and bandwidth of the preamplifier 402 to be adjusted more easily.

The switches SZ 1 and SZ 2 are utilized for cancelling a DC offset voltage across the differential input terminals of the preamplifier 402 , where the switch SZ 1 is coupled between the positive input terminal and the negative output terminal of the preamplifier 402 , and the switch SZ 2 is coupled between the negative input terminal and the positive output terminal of the preamplifier 402 . The switch S RST is coupled between the positive output terminal and the negative output terminal of the preamplifier 402 , and is utilized for resetting the differential output voltages of the preamplifier 402 . The latch circuit 404 is coupled to the positive output terminal and the negative output terminal of the preamplifier 402 , and is utilized for latching the differential output voltages of the preamplifier 402 .

Please refer to FIG. 5 , which is a schematic diagram of a subsequent stage 50 of a pipelined ADC according to an embodiment of the present invention. The subsequent stage 50 is the first subsequent stage 50 of the pipelined ADC, comprising a 1.5-bit sub-ADC 500 and a multiplying DAC 510 . The sub-ADC 500 comprises comparators 501 , 502 , and a logic circuit 504 . The comparators 501 and 502 are similar to the comparator 400 in FIG. 4 with inverse reference voltage connection. In the following description, only the comparators 501 is described since those skilled in the art can understand how the comparators operate when one of them is already known. The logic circuit 504 is coupled to the comparators 501 and 502 , and is utilized for generating a 2-bit digital signal as 01, 10, or 00 according to the voltages latched by the latch circuits of the comparators 501 and 502 .

The multiplying DAC 510 comprises a DAC 512 , an amplifier 514 , switches SM 1 -SM 5 , capacitors CM 1 , and CM 2 . Note that, the multiplying DAC 510 shown in FIG. 5 is a single-ended architecture for a simplified presentation, where the multiplying DAC 510 is a fully differential architecture, in fact. The switches SM 1 , SM 2 , the capacitors CM 1 , and CM 2 forms a switched capacitor circuit operating as a sampling circuit, where the switch SM 1 is coupled to the capacitor CM 1 , and the switch SM 2 is coupled to the capacitor CM 2 . The DAC 512 is coupled to the logic circuit 504 and the capacitor CM 2 , and is utilized for outputting the voltages V REFP , V REFN , or V CM to the capacitor CM 2 according to the 2-bit digital signal generated by the logic circuit 504 . The amplifier 514 comprises a positive input terminal coupled to the capacitors CM 1 and CM 2 , a negative input terminal coupled to the common mode voltage V CM , and an output terminal coupled to comparators in a sub-ADC in a next subsequent stage, wherein output voltages of the amplifier 514 are input voltages of the comparators in the next subsequent stage. The switch SM 3 is coupled between the output terminal of the amplifier 514 and the capacitor CM 1 . The switch SM 4 is coupled between the positive input terminal and the negative input terminal of the amplifier 514 . The switch SM 5 is coupled between the output terminal of the amplifier 514 and the common mode voltage V CM .

The clock signals used in the pipelined ADC are denoted as Φ 1 , Φ 2 , Φ 1d , Φ 2d , and Φ 2c , as shown in FIG. 4 and FIG. 5 , and are generated by a clock generator of the pipelined ADC. The clock signals Φ 1 and Φ 2 are non-overlapping sampling signals for two adjacent subsequent stage. The rising edge of the clock signals Φ 1d and Φ 2d are the same time as the rising edge of the clock signals Φ 1 and Φ 2 , and the falling edge of the clock signals Φ 1d and Φ 2d is later than the falling edge of the clock signal Φ 1 and Φ 2 . The clock signal Φ 2c is a delay clock of the clock signal Φ 2d . Note that, the way of the switches being coupled to the reference voltages and the input voltages according to the present invention is different from that in the conventional switched capacitor circuit, and the way of using the clock signals is different, so that the comparator according to the present invention brings effects different from the conventional comparator.

Please refer to FIG. 4 and FIG. 5 . When the clock signal Φ 2c is at a high voltage level, the switches S 1 A, S 2 A, S 3 A, and S 4 A are turned on and other switches are turned off, so that the positive reference voltage V REFP is sampled to the capacitors C 1 , Cs 1 , C 3 , and Cs 3 , and the negative reference voltage V REFN is sampled to the capacitors C 2 , Cs 2 , C 4 , and Cs 4 . At the same time, the switches SZ 1 and SZ 2 are turned on, so that the DC offset voltage across the differential input terminals of the preamplifier 402 is zeroed. The logic circuit 504 outputs the 2-bit digital signal 00, 01, or 10 to the DAC 512 according to the voltages latched by the latch circuits of the comparators 501 and 502 at the rising edge of the clock signal Φ 2c . The DAC 512 outputs a corresponding one of the voltages V REFP , V REFN , or V CM to the capacitor CM 2 according to the 2-bit digital signal. When the clock signal Φ 2c is at a high voltage level, the switch SM 3 is turned on, and the capacitor CM 1 and the amplifier 514 form a negative feedback path. After the input voltage V IN is amplified, the output voltage of the DAC 512 is subtracted from the amplified input voltage, and then a residue voltage is generated, outputted to a next subsequent stage.

When the clock signal Φ 1 is at a high voltage level, the switches S 1 B, S 2 B, S 3 B, and S 4 B are turned on and other switches are turned off, so that the positive input voltage V INP is sampled to the capacitors Cs 1 , Cs 2 , and a serial capacitor formed by the capacitors C 1 , C 2 , and Cp 1 , and similarly, the negative input voltage V INN is sampled to the capacitors Cs 3 , Cs 4 , and a serial capacitor formed by the capacitors C 3 , C 4 , and Cp 2 . At the same time, the switch S RST is turned on to reset the differential output voltages of the preamplifier 402 , and therefore the recovery time of the comparator 400 is reduced. For the multiplying DAC 510 , when the clock signal Φ 1 is at a high voltage level, the switches SM 4 and SM 5 are turned on, so that the differential input terminals of the amplifier 514 are pulled to the common mode voltage V CM . In addition, when the clock signal Φ 1d is at a high voltage level, the switches SM 1 and SM 2 are turned on, so that the input voltage of the multiplying DAC 510 are sampled to the capacitors CM 1 and CM 2 .

At the falling edge of the clock signal Φ 1 , the voltage difference (V INP −V INN ) is compared to the threshold +¼ (V REFP −V REFN ) and −¼ (V REFP −V REFN ). That is, the sampled voltage difference (V INP −V INN ) is quantized with two threshold level. The preamplifier 402 amplifies the voltage across the differential input terminals and outputs an amplified voltage across the differential output terminals, which indicates a comparison result, to the latch circuit 404 . The latch circuit 404 latches the differential output voltages of the preamplifier 402 at the rising edge of the clock signal Φ 2c . From the above, it is known that the preamplifier 402 performs amplification when a time between the falling edge of the clock signal Φ 1 and the rising edge of the clock signal Φ 2c , as T d1 shown in FIG. 4 . As shown in FIG. 3 , in the conventional comparator 202 , the preamplifier 206 performs amplification only when a time T d between the rising edge of the clock signal Φ 2d and the rising edge of the clock signal Φ 2c . In comparison, the preamplifier 402 has the operation time T d1 longer than T d . Therefore, the preamplifier 402 has more time to amplify the voltage difference across the differential input terminals to reach an enough voltage level, so that the logic circuit has less opportunity to output a wrong signal.

Furthermore, to avoid aperture error, the switched capacitor circuit of the comparator 501 (or 502 ) and the switched capacitor circuit of the multiplying DAC 510 should provide the same time constant. Assume that the capacitances of the capacitors Cs 1 -Cs 4 are equal, denoted as Cs, the capacitances of the capacitors Cp 1 and Cp 2 are equal, denoted as Cp, the capacitances of the capacitors C 1 and C 4 are equal, the capacitances of the capacitors C 2 and C 3 are equal, and the sum of the capacitances of C 1 and C 2 is far large than the capacitance Cp, the resistances and capacitances used in the switched capacitor circuit need to be fulfilled as the following:

R M ⁢ C M = R C ⁡ ( 2 ⁢ ⁢ C S + ( C 1 + C 2 ) ⁢ C p C 1 + C 2 + C p ) ≈ R C ⁡ ( 2 ⁢ ⁢ C S + C p ) , ( 2 )

when the capacitance Cs is far large than the capacitance Cp, the equation 2 is simplified as:

R M C M ≈2R C C S ,  (3)

where R M is the equivalent resistance of the switch SM 1 or SM 2 in the multiplying DAC 510 , C M is the capacitance of the capacitor CM 1 or CM 2 , and R C is the equivalent resistance of the switch in the switched capacitor circuit of the comparator 501 or 502 . From the above, the present invention uses a clock signal control scheme different from the prior art, and uses the additional sampling capacitors Cs 1 -Cs 4 , to control the time constant of the switched capacitor circuit of the comparator to be not effected by the transconductance of the transistor in the preamplifier. According to the present invention, when the capacitors and the switches in the switched capacitor circuit are properly designed, the time constants for the comparators 501 and for the multiplying DAC 510 can be easily matched without adjusting the transconductance of the transistor in the preamplifier, and thereby the aperture error is not generated.

Briefly, when the first subsequent stage of a pipelined ADC uses the comparator 400 , a front-end sample-and-hold amplifier in front of the first subsequent stage is not required. The time for the preamplifier 402 to perform amplification is longer than that for the conventional preamplifier, so that the possibility of the logic circuit outputting a wrong signal is reduced. The time constant of the switched capacitor circuit of the comparator 400 is related to the capacitance and the equivalent resistance of the switch, and is not related to the transconductance of the transistor, which brings an easy way to control the aperture error. In addition, the reference voltage generator of the pipelined ADC does not require generating additional reference voltages for the comparator 400 as for the conventional comparator in FIG. 3 , and therefore power consumption of the pipelined ADC is reduced.

Please refer to FIG. 6 , which is a schematic diagram of a comparator 600 according to an embodiment of the present invention. The comparator 600 is used in a subsequent stage other than the first subsequent stage of a pipelined ADC. The comparator 600 comprises a preamplifier 602 , a latch circuit 604 , and a switched capacitor circuit 606 . The switched capacitor circuit 606 comprises switches S 1 A, S 1 B, S 2 A, S 2 B, S 3 A, S 3 B, S 4 A, S 4 B, SZ 1 , SZ 2 , and capacitors C 1 -C 4 . Compared to the comparator 400 in FIG. 4 , the comparator 600 does not comprise the sampling capacitors Cs 1 -Cs 4 and the switch S RST because input voltages of any other subsequent stage behind the first subsequent stage are close to DC voltages, so that sampling is not required. Except the capacitors Cs 1 -Cs 4 and the switch S RST , the comparator 600 is similar to the comparator 400 . Please refer to the aforementioned embodiments to realize the components in the comparator 600 .

Note that, the comparator 600 can be utilized in the different subsequent stage, e.g. the second or the third subsequent stage, with different clock signals. Please refer to FIG. 7 and FIG. 8 , which are schematic diagrams of the second subsequent stage 70 and the third subsequent stage 80 behind the first subsequent stage 50 in FIG. 5 . The second subsequent stage 70 or the third subsequent stage 80 includes a sub-ADC including the comparator 600 in FIG. 6 and the logic circuit 504 , and the multiplying DAC 510 . Note that, in FIG. 7 and FIG. 8 only one comparator is illustrated for a simplified presentation. As shown in FIG. 7 , in the subsequent stage 70 , the clock signal Φ 2 controls the switches in the comparator 600 to sample the differential input voltages V INP and V INN . The clock signal Φ 2d controls the switches in the multiplying DAC 510 to sample the input voltage V IN . The clock signal Φ 1d control the switches in the comparator 600 to sample the reference voltages V REFP and V REFN . The latch circuit 604 latches the differential output voltages of the preamplifier 602 at the rising edge of the clock signal Φ 1d .

As shown in FIG. 8 , in the third subsequent stage 80 , the clock signal Φ 1 controls the switches in the comparator 600 to sample the differential input voltages V INP and V INN . The clock signal Φ 1d controls the switches in the multiplying DAC 510 to sample the input voltage V IN . The clock signal Φ 2d control the switches in the comparator 600 to sample the reference voltages V REFP and V REFN . The latch circuit 604 latches the differential output voltages of the preamplifier 602 at the rising edge of the clock signal Φ 2d . Those skilled in the art can know how any other subsequent stage behind the third subsequent stage operates according to the clock signals Φ 1 , Φ 1d , Φ 2 , and Φ 2d , which are omitted herein.

For implementing the comparator 400 , not only the components in the comparator 400 but also the clock signals Φ 1 , Φ 1d , Φ 2 , Φ 2d , and Φ 2c are required. Please refer to FIG. 9 , which is a flowchart of a process 90 according to an embodiment of the present invention. The process 90 illustrates operation of the switched capacitor circuit in the comparator 400 , which comprises the following steps:

Step 902 : Generate clock signals Φ 1 , Φ 1d , Φ 2 , Φ 2d and Φ 2c .

Step 904 : Control the switches S 1 B, S 2 B, S 3 B, and S 4 B according to the clock signal Φ 1 to sample the positive input voltage V INP and the negative input voltage V INN of the comparator 400 on the capacitors C 1 -C 4 and Cs 1 -Cs 4 .

Step 906 : Control the switches S 1 A, S 2 A, S 3 A, and S 4 A according to the clock signal Φ 2c to sample the positive reference voltage V REFP and the negative reference voltage V REFN on the capacitors C 1 -C 4 and Cs 1 -Cs 4 .

Please refer to the aforementioned embodiments to realize how the process 90 is performed. Through the process 90 , the switched capacitor circuit of the comparator 400 can sample the differential input voltages and differential reference voltages on the capacitors at different time, so as to make the time constant for the comparator 400 only related to the equivalent resistance of the switch and the capacitance, and not related to the transconductance of the transistor Therefore, the aperture error is avoided.

In conclusion, when the first subsequent stage of a pipelined ADC applies the comparator and the process according to the present invention, the pipelined ADC does not require a front-end sample-and-hold amplifier, cost and power consumption of the pipelined ADC are therefore reduced. Preferably, the comparator according to the present invention provides a longer time for signal amplification and a time constant easy to be adjusted, so that the aperture error is obviously reduced. Compared with the conventional comparator, the comparator according to the present invention is more useful in high-speed application.

Those skilled in the art will readily observe that numerous modifications and alterations of the device and method may be made while retaining the teachings of the invention.