US20100321220A1 - 1-bit cell circuit used in a pipelined analog to digital converter - Google Patents
1-bit cell circuit used in a pipelined analog to digital converter Download PDFInfo
- Publication number
- US20100321220A1 US20100321220A1 US12/489,778 US48977809A US2010321220A1 US 20100321220 A1 US20100321220 A1 US 20100321220A1 US 48977809 A US48977809 A US 48977809A US 2010321220 A1 US2010321220 A1 US 2010321220A1
- Authority
- US
- United States
- Prior art keywords
- voltage
- input
- plate
- negative
- positive
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012546 transfer Methods 0.000 claims abstract description 41
- 239000003990 capacitor Substances 0.000 claims description 140
- 238000005070 sampling Methods 0.000 claims description 60
- 238000010586 diagram Methods 0.000 description 14
- 238000000034 method Methods 0.000 description 3
- 230000000630 rising effect Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000012552 review Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M1/00—Analogue/digital conversion; Digital/analogue conversion
- H03M1/12—Analogue/digital converters
- H03M1/34—Analogue value compared with reference values
- H03M1/38—Analogue value compared with reference values sequentially only, e.g. successive approximation type
- H03M1/44—Sequential comparisons in series-connected stages with change in value of analogue signal
- H03M1/442—Sequential comparisons in series-connected stages with change in value of analogue signal using switched capacitors
Definitions
- the present invention relates to pipelined ADCs, and more particularly to a 1 -bit cell circuit used in pipelined ADCs.
- FIG. 1 shows the typical architecture of a prior art 1-bit cell of a pipelined ADC.
- the 1-bit cell has a sampling phase and a charge transfer phase during a clock cycle.
- the 1-bit cell includes an OTA (Operational Transconductance Amplifier) 101 , four matched capacitors 102 ⁇ 105 , a first set of switches 106 ⁇ 113 , a second set of switches 114 ⁇ 121 , two multiplexers 122 and 123 , and two latched comparators 124 and 125 .
- OTA Operaational Transconductance Amplifier
- the OTA 101 is used to generate a residue output Vout according to an input signal Vin, wherein the residue output Vout is composed of a positive output Voutp and a negative output Voutn, and the input signal Vin is composed of a positive input signal Vinp and a negative input signal Vinn.
- the matched capacitors 102 ⁇ 105 are used to hold a sampled voltage of the input signal Vin.
- the first set of switches 106 ⁇ 113 are closed for sampling the input signal Vin during the sampling phase.
- the second of switches 114 ⁇ 121 are closed during the charge transfer phase to make the matched capacitors 102 ⁇ 105 and the OTA 101 form a negative feedback circuit to generate the residue output Vout.
- the two multiplexers 122 and 123 are used to respectively provide a first multiplexer output voltage and a second multiplexer output voltage which are selected from a group consisting of a negative reference voltage Vrefn, a positive reference voltage Vrefp and a ground voltage according to two select signals bp and bn, wherein bp is a positive bit signal and bn is a negative bit signal.
- the comparator 124 is used to generate the positive bit signal bp according to voltage comparison of the input signal Vin and a first reference voltage Vref/ 4
- the comparator 125 is used to generate the negative bit signal bn according to voltage comparison of the input signal Vin and a second reference voltage ⁇ Vref/4, wherein the Vref is equal to the voltage difference between the Vrefp and the Vrefn.
- FIG. 2 a shows an illustrating diagram of the sampling phase of the prior art 1-bit cell.
- the top plates of the capacitor 102 and the capacitor 103 are connected to the positive terminal of the Vin
- the top plates of the capacitor 104 and the capacitor 105 are connected to the negative terminal of the Vin
- all the bottom plates of the capacitors 102 , 103 , 104 and 105 are connected to the ground.
- the comparator 124 and 125 respectively generates the positive bit signal bp and the negative bit signal bn for the charge transfer phase.
- FIG. 2 b shows an illustrating diagram of the charge transfer phase of the prior art 1-bit cell.
- Vref the residue output
- the total amount of charge Q taken from the Vref by each 1-bit cell can be easily computed as being the capacitance of the capacitor 102 , the capacitor 104 times the voltage difference between the final voltage appearing on the capacitor 102 , the capacitor 104 at the end of the charge transfer phase, and the voltage appearing on the capacitor 102 , the capacitor 104 at the beginning of the charge transfer phase.
- the total amount of charge Q is derived according to bit decision as follows:
- the present invention proposes a novel topology of signal independent voltage reference loading for the pipelined ADC.
- One objective of the present invention is to provide a signal independent voltage reference loading circuit to construct a concise and power saving pipelined ADC.
- Another objective of the present invention is to provide a signal independent voltage reference loading circuit to release the speed requirement of the voltage reference buffer.
- Still another objective of the present invention is to provide a signal independent voltage reference loading circuit to construct a 4 x interleaved pipelined ADC.
- the present invention provides a 1-bit cell circuit used in a pipelined analog to digital converter, having a sampling phase including a first half period and a second half period and a charge transfer phase in a cycle.
- the 1-bit cell circuit comprises a differential amplifier, having a positive input end, a negative input end, a positive output end and a negative output end, for outputting a residue output signal; a first capacitor, having a first plate and a second plate, and a second capacitor, having a third plate and a fourth plate, wherein said first plate is connected to a positive input signal, said third plate is connected to a negative input signal, and said second plate and said fourth plate are both connected to an input common mode reference ground during said sampling phase, and said first plate is connected to said third plate, said second plate is connected to said negative input end of said differential amplifier, and said fourth plate is connected to said positive input end of said differential amplifier during said charge transfer phase; and a third capacitor, having a fifth plate and a sixth plate, and a fourth capacitor, having
- a 1-bit cell circuit is used in a pipelined analog to digital converter, having a sampling phase including a first half period and a second half period and a charge transfer phase in a cycle.
- the 1-bit cell circuit comprises a differential amplifier, having a pair of differential input ends and a pair of differential output ends, for outputting a differential residue output signal; a first pair of capacitors, having a pair of first top plates and a pair of first bottom plates, wherein said pair of first top plates are connected to a differential input signal, and said first bottom plates are connected to an input common mode reference ground during said sampling phase, and said first top plates are connected to each other and said first bottom plates are connected to said differential input ends of said differential amplifier during said charge transfer phase; and a second pair of capacitors, having a pair of second top plates and a pair of second bottom plates, wherein said second top plates are connected to an output common mode reference ground and said second bottom plates are connected to an input common mode reference ground during said first half period of said sampling phase; said
- a 1-bit cell circuit is used in a pipelined analog to digital converter.
- the 1-bit cell circuit comprises a reference buffer for providing a reference voltage; a sample and charge transfer circuit for receiving an input signal to generate an output signal; and a dump circuit for dumping said reference voltage; wherein said reference buffer selectively connects to one of said sample and charge transfer circuit and said dump circuit according to said input signal.
- FIG. 1 is the typical architecture of a prior art 1-bit cell of a pipelined ADC.
- FIG. 2 a is an illustrating diagram of sampling phase of the prior art 1-bit cell.
- FIG. 2 b is an illustrating diagram of charge transfer phase of the prior art 1-bit cell.
- FIG. 3 is the circuit diagram of a 1-bit cell using a signal independent voltage reference loading circuit according to a preferred embodiment of the present invention.
- FIG. 4 a is an illustrating diagram of a first half period of sampling phase of the 1-bit cell in FIG. 3 .
- FIG. 4 b is an illustrating diagram of a second half period of sampling phase of the 1-bit cell in FIG. 3 .
- FIG. 4 c is an illustrating diagram of a charge transfer phase of the 1-bit cell in FIG. 3 .
- FIG. 5 is a circuit diagram of a cascade of two 1-bit cells according to a preferred embodiment of the present invention.
- FIG. 3 a ⁇ b shows the circuit diagram of a 1-bit cell using a signal independent voltage reference loading circuit according to a preferred embodiment of the present invention.
- the 1-bit cell ADC procedure includes a first half period of sampling phase, a second half period of sampling phase and a charge transfer phase during a converting period. As shown in FIG.
- the 1-bit cell ADC serve as a sample and charge transfer circuit
- OTA Operaational Transconductance Amplifier
- six matched capacitors 302 ⁇ 307 eight sampling phase switches 308 ⁇ 315 , seven charge transfer phase switches 316 ⁇ 322 , two multiplexers 323 and 324 , two comparators 337 and 338 and a voltage reference buffer 339 .
- a dump circuit comprises two multiplexers 330 and 336 , two matched capacitors 325 and 331 , four sampling phase switches 326 ⁇ 327 and 332 ⁇ 333 , and four charge transfer phase switches 328 ⁇ 329 and 334 ⁇ 335 .
- the OTA 301 having a positive input end, a negative input end, a positive output end and a negative output end, is used to generate a residue output V out according to an input signal V in , wherein the residue output V out is composed of a positive output V outp and a negative output V outn , and the input signal V in is composed of a positive input signal V inp and a negative input signal V inn .
- the eight matched capacitors 302 ⁇ 307 , 325 and 331 are implemented to function as follows. In the circuit, the capacitors 302 , 303 can be viewed as a first capacitor and the capacitors 305 , 306 can be viewed as a second capacitor.
- the first capacitor and the second capacitor can be viewed as a first pair of capacitors; the capacitors 304 , 307 can be viewed as a second pair of capacitors; and the capacitors 325 , 331 can be viewed as a third pair of capacitors, wherein the first pair of capacitors are used to hold the sampled voltage of the input signal, and the second pair of capacitors are used to hold the sampled voltage of a differential voltage, which is selected from one of a reference voltage and an output common mode reference ground, and the third pair of capacitors are used to serve as a dump circuit for the reference voltage when the reference voltage is not selected to charge the second pair of capacitors.
- the first capacitor composed of capacitors 302 and 303 , both implemented according to a predetermined capacitance, is used to hold a sampled voltage of the positive input signal V inp .
- the capacitor 304 implemented according to the predetermined capacitance, is used to hold one voltage selected from a group consisting of a positive reference voltage V refp , a negative reference voltage V refn and the output common mode reference ground.
- the second capacitor composed of capacitors 305 and 306 , both implemented according to the predetermined capacitance, is used to hold a sampled voltage of the negative input signal V inn .
- the capacitor 307 implemented according to the predetermined capacitance, is used to hold one voltage selected from a group consisting of the positive reference voltage V refp , the negative reference voltage V refn and the output common mode reference ground.
- the capacitor 325 implemented according to the predetermined capacitance, is used to provide a dump load for the positive reference voltage V refp when the output voltage of the multiplexer 323 and the output voltage of the multiplexer 324 both connect to the output common mode reference ground.
- the capacitor 331 implemented according to the predetermined capacitance, is used to provide a dump load for the negative reference voltage V refn when the output voltage of the multiplexer 323 and the output voltage of the multiplexer 324 both connect to the output common mode reference ground.
- the twelve sampling phase switches 308 ⁇ 315 , 326 ⁇ 327 and 332 ⁇ 333 are implemented to function during the sampling phase as follows.
- the switches 308 and 309 are used to sample the positive input signal V inp into the capacitors 302 and 303 .
- the switch 310 is used to connect the bottom plate of the capacitor 304 to an input common mode reference ground and the switch 311 is used to connect the top plate of the capacitor 304 to the output of the multiplexer 323 .
- the switches 312 and 313 are used to sample the negative input signal V inn into the capacitors 305 and 306 .
- the switch 314 is used to connect the bottom plate of the capacitor 307 to the input common mode reference ground and the switch 315 is used to connect the top plate of the capacitor 307 to the output of the multiplexer 324 .
- the switch 326 is used to connect the bottom plate of the capacitor 325 to the input common mode reference ground and the switch 327 is used to connect the top plate of the capacitor 325 to the output of the multiplexer 330 .
- the switch 332 is used to connect the bottom plate of the capacitor 331 to the input common mode reference ground and the switch 333 is used to connect the top plate of the capacitor 331 to the output of the multiplexer 336 .
- the eleven charge transfer phase switches 316 ⁇ 322 , 328 ⁇ 329 and 334 ⁇ 335 are implemented to function during the charge transfer phase as follows.
- the switch 316 is used to interconnect the top plates of the capacitors 302 , 303 , 305 and 306 .
- the switch 317 is used to connect the bottom plates of the capacitors 302 , 303 to the negative input end of the OTA 301 .
- the switch 318 is used to connect the bottom plate of the capacitor 304 to the negative input end of the OTA 301 .
- the switch 319 is used to connect the bottom plate of the capacitors 304 to the positive output end of the OTA 301 .
- the switch 320 is used to connect the bottom plates of the capacitors 305 and 306 to the positive input end of the OTA 301 .
- the switch 321 is used to connect the bottom plate of the capacitor 307 to the positive input end of the OTA 301 .
- the switch 322 is used to connect the top plate of the capacitor 307 to the negative output end of the OTA 301 .
- the capacitors 302 ⁇ 307 and the OTA 301 together form a negative feedback circuit to generate the residue output V out during the charge transfer phase.
- the switch 328 is used to connect the bottom plate of the capacitor 325 to the input common mode reference ground and the switch 329 is used to connect the top plate of the capacitor 325 to the output common mode reference ground.
- the switch 334 is used to connect the bottom plate of the capacitor 331 to the input common mode reference ground and the switch 335 is used to connect the top plate of the capacitor 331 to the output common mode reference ground.
- the multiplexer 323 is used to provide a first multiplexer output voltage selected from a group consisting of the negative reference voltage V refn , the positive reference voltage V refp and the output common mode reference ground according to two select signals bp&e and bn&e, wherein bp&e is logic-AND of a positive bit signal bp and a clock signal e, and bn&e is logic-AND of a negative bit signal bn and the clock signal e, and the rising edge of the clock e is the breaking point of the first half period and the second half period of sampling phase.
- the multiplexer 324 is used to provide a second multiplexer output voltage selected from a group consisting of the negative reference voltage V refn , the positive reference voltage V refp and the output common mode reference ground according to bp&e and bn&e.
- the second multiplexer output voltage is connected to the output common mode reference ground during the first half period of sampling phase, and connected to one of the negative reference voltage V refn , the positive reference voltage V refp and the output common mode reference ground according to the bp&e and bn&e during the second half period of sampling phase.
- the multiplexer 330 is used to provide a third multiplexer output voltage selected from a group consisting of the positive reference voltage V refp and the output common mode reference ground according to a select signal e& ⁇ (bp+bn).
- the third multiplexer output voltage is connected to the output common mode reference ground during the first half period of sampling phase, and connected to one of the positive reference voltage V refp and the output common mode reference ground according to the e& ⁇ (bp+bn) during the second half period of sampling phase.
- the fourth multiplexer output voltage is connected to the output common mode reference ground during the first half period of sampling phase, and connected to one of the negative reference voltage V refp and the output common mode reference ground according to the e& ⁇ (bp+bn) during the second half period of sampling phase.
- the comparator 337 having a first positive comparing input end, a first negative comparing input end and a first comparing output end, is used to generate the positive bit signal bp according to voltage comparison of the input signal V in and a first reference voltage V ref /4 at the rising edge of the clock e, wherein the V ref is equal to the voltage difference between the V refp and the V refn .
- the comparator 338 having a second positive comparing input end, a second negative comparing input end and a second comparing output end, is used to generate the negative bit signal bn according to voltage comparison of the input signal V in and a second reference voltage ⁇ V ref /4 at the rising edge of the clock e, wherein (bp, bn) is (0,0) when V in is between ⁇ V ref /4 and V ref /4; (bp, bn) is (0,1) when V in is smaller than ⁇ V ref /4; and (bp, bn) is (1,0) when V in is greater than V ref /4.
- the voltage reference buffer 339 is used to provide the V refp and the V refn , and the voltage reference buffer 339 can be an OTA or a low output impedance amplifier.
- FIG. 4 a shows an illustrating diagram of the first half period of sampling phase of the 1-bit cell in FIG. 3 .
- the top plates of the capacitors 302 and 303 are connected to the positive terminal of the V in
- the top plates of the capacitors 305 and 306 are connected to the negative terminal of the V in
- all the bottom plates of the capacitors 302 , 303 , 305 and 306 are connected to the input common mode reference ground.
- the bottom plates of the capacitors 304 , 307 are connected to the input common mode reference ground, and the top plate of the capacitor 304 is connected to a first reference contact, which is the output of the multiplexer 323 , being connecting to the output common mode reference ground according to the bp&e and bn&e during the first half period of sampling phase, and the top plate of the capacitor 307 is connected to a second reference contact, which is the output of the multiplexer 324 , being connecting to the output common mode reference ground according to the bp&e and bn&e during the first half period of sampling phase.
- the bottom plates of the capacitor 325 , 331 are connected to the input common mode reference ground, and the top plates of the capacitors 325 , 331 are respectively connected to the output ends of the multiplexer 330 , 336 , being both connecting to the output common mode reference ground according to the e& ⁇ (bp+bn).
- the comparator 337 generates the positive bit signal bp and the comparator 338 generates the negative bit signal bn at the end of the first half period of sampling phase.
- FIG. 4 b shows an illustrating diagram of the second half period of sampling phase of the 1-bit cell in FIG. 3 .
- the top plates of the capacitors 302 and 303 are connected to the positive terminal of the V in
- the top plates of the capacitors 305 and 306 are connected to the negative terminal of the V in
- all the bottom plates of the capacitors 302 , 303 , 305 and 306 are connected to the input common mode reference ground.
- the top plate of the capacitor 304 is connected to the first reference contact, which is the output of the multiplexer 323 , being connecting to one of the negative reference voltage V refn , the positive reference voltage V refp and the output common mode reference ground according to the bp&e and bn&e.
- the top plate of the capacitor 307 is connected to the second reference contact, which is the output of the multiplexer 324 , being connecting to one of the negative reference voltage V refn , the positive reference voltage V refp and the output common mode reference ground according to the bp&e and bn&e.
- the voltage difference between the top plates of the capacitor 304 and 307 is therefore equal to a first differential voltage (bn-bp)*V ref .
- the bottom plate of the capacitor 325 is connected to the input common mode reference ground and the top plate of the capacitors 325 is connected to the output of the multiplexer 330 , which is connected to one of the positive reference voltage V refp and the output common mode reference ground according to the e& ⁇ (bp+bn).
- the bottom plate of the capacitor 331 is connected to the input common mode reference ground and the top plate of the capacitors 331 is connected to the output of the multiplexer 336 , which is connected to one of the negative reference voltage V refn and the output common mode reference ground according to the e& ⁇ (bp+bn).
- the voltage difference between the top plates of the capacitor 325 and 331 is therefore equal to a second differential voltage (1-bn-bp)*V ref . It can be proven that the sum of the absolute value of the first differential voltage (bn-bp)*V ref and the absolute value of the second differential voltage (1-bn-bp)*V ref is equal to V ref no matter what the (bp,bn) may be.
- FIG. 4 c shows an illustrating diagram of the charge transfer phase of the 1-bit cell in FIG. 3 .
- the top plates of the capacitors 302 ⁇ 303 , 305 ⁇ 306 are connected together, and the bottom plates of the capacitors 302 ⁇ 303 are connected to the negative input end of the OTA 301 , and the bottom plates of the capacitors 305 ⁇ 306 are connected to the positive input end of the OTA 301 ;
- the bottom plate of the capacitor 304 is connected to the negative input end of the OTA 301 , and the top plate of the capacitor 304 is connected to the positive output end of the OTA 301 , and the bottom plate of the capacitor 307 is connected to the positive input end of the OTA 301 , and the top plate of the capacitor 307 is connected to the negative output end of the OTA 301 ;
- the bottom plates of the capacitor 325 , 331 are connected to the input common mode reference ground and the top plates of the capacitor 325 , 331 are connected to the output common mode reference ground
- FIG. 5 shows a circuit diagram of a cascade of two 1-bit cells in a pipelined ADC according to a preferred embodiment of the present invention.
- the residue output of a 1-bit cell 510 is connected to the input of a 1-bit cell 520 , and the 1-bit cell 510 generates a positive bit signal bp 1 and a negative bit signal bn 1 , and the 1-bit cell 520 generates a positive bit signal bp 2 and a negative bit signal bn 2 .
- V ref is not used at all, i.e., not used by the cell 510 (and all other odd cells not depicted), nor by the cell 520 (and all other even cells not depicted). This allows the reference voltage to be shared with another set of cells operated with half a clock phase delay, to build a 4 ⁇ time-interleaved ADC.
- the positive reference voltage V refp has a load capacitor with constant initial charge no matter what bp and bn may be during the second half period of sampling phase
- the negative reference voltage V refn has a load capacitor with constant initial charge no matter what bp and bn may be during the second half period of sampling phase
- the sum of the absolute value of a first differential voltage, which is the voltage difference between the first multiplexer output voltage and the second multiplexer output voltage, and the absolute value of a second differential voltage, which is the voltage difference between the third multiplexer output voltage and the fourth multiplexer output voltage, is equal to a settling voltage of the reference voltage V ref at the end of the second half period of sampling phase, wherein the reference voltage V ref is the voltage difference between the positive reference voltage V refp and the negative reference voltage V refn .
- the novel 1-bit cell ensures that the amount of charges provided by the reference buffer in each clock period is not dependent on the signal currently converted. Besides, assuming that the clock has achieved the minimal jitter required for aperture noise concerns on high frequency input signal, the novel 1-bit cell ensures both a constant load and a constant settling time for the reference buffer to have a constant settling, even if incomplete, whereas, with a load depending on input signal, full settling is the only way to guaranty a constant final value.
- the group delay of a 1-bit cell is critical for the data rate of a pipelined converter in that a shorter group delay will result in a higher data rate, and also critical for the total conversion delay, the time elapsed between signal sampling and full n-bit code delivery, which is of extreme importance in some application like AGC (Automatic Gain Control) or other closed loop systems.
- the 1-bit cell of the present invention preserves the fundamental 1-clock phase group delay, i.e., a new residue output is delivered to the next cell exactly one clock phase after an old residue was sampled from the previous cell, so it can offer a smallest group delay, a smallest total conversion delay and therefore a high data rate.
- the present invention thus possesses the following advantages.
- the settling voltage of the second pair of capacitors needs not equal to the reference voltage V ref , so the speed requirement on the reference buffer can be relaxed, and the power needed by the reference buffer is thereby reduced.
- the voltage reference V ref since during the second half period of sampling phase the voltage reference V ref always has a pair of capacitors implemented according to a predetermined capacitance, and the pair of capacitors being with constant initial charge, the total charge supplied by the voltage reference V ref is therefore constant, i.e. independent of the input signal. Even though there exists mismatch between the capacitors, for example, ⁇ 0.1%, the variation of the charge taken from the voltage reference V ref will be ⁇ 0.1% ⁇ the predetermined capacitance ⁇ V ref /2, much smaller than the 0.75 ⁇ Cs ⁇ V ref /2 of the prior art.
- the settling of the reference buffer is not interfering with the setting of the OTA's (no cascade setting), and thereby avoids any slow down of each cell settling.
- the reference buffer since the reference buffer is only used half the time, it can, within same power, be shared by one more ADC time-shifted by half a clock cycle to build a 4 ⁇ interleaved ADC even if scaling down design is progressively applied from 1 st (MSB) to last (LSB) cells for optimal power and area concerns.
- MSB 1 st
- LSB last
- the present invention herein enhances the performance than the conventional structure and further complies with the patent application requirements and is submitted to the Patent and Trademark Office for review and granting of the commensurate patent rights.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Analogue/Digital Conversion (AREA)
Abstract
Description
- 1. Field of the Invention
- The present invention relates to pipelined ADCs, and more particularly to a 1-bit cell circuit used in pipelined ADCs.
- 2. Description of the Related Art
- In mixed modes circuits, ADC (Analog-Digital-Converter) is an inevitable part. Recently, pipelined ADC is widely adopted due to its concise structure and superior performance. Please refer to
FIG. 1 , which shows the typical architecture of a prior art 1-bit cell of a pipelined ADC. The 1-bit cell has a sampling phase and a charge transfer phase during a clock cycle. As shown inFIG. 1 , the 1-bit cell includes an OTA (Operational Transconductance Amplifier) 101, four matchedcapacitors 102˜105, a first set ofswitches 106˜113, a second set ofswitches 114˜121, two 122 and 123, and twomultiplexers 124 and 125.latched comparators - The
OTA 101 is used to generate a residue output Vout according to an input signal Vin, wherein the residue output Vout is composed of a positive output Voutp and a negative output Voutn, and the input signal Vin is composed of a positive input signal Vinp and a negative input signal Vinn. The matchedcapacitors 102˜105 are used to hold a sampled voltage of the input signal Vin. The first set ofswitches 106˜113 are closed for sampling the input signal Vin during the sampling phase. The second ofswitches 114˜121 are closed during the charge transfer phase to make the matchedcapacitors 102˜105 and theOTA 101 form a negative feedback circuit to generate the residue output Vout. The two 122 and 123 are used to respectively provide a first multiplexer output voltage and a second multiplexer output voltage which are selected from a group consisting of a negative reference voltage Vrefn, a positive reference voltage Vrefp and a ground voltage according to two select signals bp and bn, wherein bp is a positive bit signal and bn is a negative bit signal. When bp=0 and bn=0, the first multiplexer output voltage and the second multiplexer output voltage are both connected to the ground voltage; when bp=0 and bn=1, the first multiplexer output voltage is connected to the negative reference voltage Vrefn and the second multiplexer output voltage is connected to the positive reference voltage Vrefp; and when bp=1 and bn=0, the first multiplexer output voltage is connected to the positive reference voltage Vrefp and the second multiplexer output voltage is connected to the negative reference voltage Vrefn. Themultiplexers comparator 124 is used to generate the positive bit signal bp according to voltage comparison of the input signal Vin and a first reference voltage Vref/4, and thecomparator 125 is used to generate the negative bit signal bn according to voltage comparison of the input signal Vin and a second reference voltage −Vref/4, wherein the Vref is equal to the voltage difference between the Vrefp and the Vrefn. When Vin is between −Vref/4 and Vref/4, (bp, bn) is (0,0); when Vin is smaller than −Vref/4, (bp, bn) is (0,1); and when Vin is greater than Vref/4, (bp, bn) is (1,0). - During the sampling phase, the input signal Vin is sampled and hold, and the positive bit signal bp and the negative bit signal bn are generated. Please refer to
FIG. 2 a, which shows an illustrating diagram of the sampling phase of the prior art 1-bit cell. As shown inFIG. 2 a, the top plates of thecapacitor 102 and thecapacitor 103 are connected to the positive terminal of the Vin, and the top plates of thecapacitor 104 and thecapacitor 105 are connected to the negative terminal of the Vin, and all the bottom plates of the 102, 103, 104 and 105 are connected to the ground. Thecapacitors 124 and 125 respectively generates the positive bit signal bp and the negative bit signal bn for the charge transfer phase.comparator - During the charge transfer phase, the residue output is generated. Please refer to
FIG. 2 b, which shows an illustrating diagram of the charge transfer phase of the prior art 1-bit cell. As shown inFIG. 2 b, due to the virtual ground of theOTA 101 in negative feedback, there will be charge drawn from the Vref, flowing through thecapacitors 102˜105 to generate the residue output Vout=2×Vin−(bp−bn)×Vref. - However, as the Vin is a varying signal, the charge delivered by the Vref will be different. That is, the loading of the Vref will be varying and dependent on the input signal Vin. The total amount of charge Q taken from the Vref by each 1-bit cell can be easily computed as being the capacitance of the
capacitor 102, thecapacitor 104 times the voltage difference between the final voltage appearing on thecapacitor 102, thecapacitor 104 at the end of the charge transfer phase, and the voltage appearing on thecapacitor 102, thecapacitor 104 at the beginning of the charge transfer phase. For a first order analysis, assuming complete settling at each clock phase, and neglecting any mismatch between the positive and negative channels of the differential architecture, e.g. consider the capacitance of thecapacitor 102=Csp, the capacitance of thecapacitor 104=Csn, and Csp=Csn=Cs. The total amount of charge Q is derived according to bit decision as follows: -
for b=+1(bp=1&bn=0), Q=Csp*(Vrefp−Vinp)=−Csn*(Vrefn−Vinn)=Cs*(Vref−Vin)/2; -
for b=−1(bp=0&bn=1), Q=Csn*(Vrefp−Vinn)=Csp(Vrefn−Vinp)=Cs*(Vref+Vin)/2; and -
for b=0(bp=0&bn+0), Q=0. - The minimal input voltage Vin yielding a positive bit decision (b=+1) is Vref/4, and the maximal amount of charge thereof arises to: Q,max=Cs*(Vref−Vref/4)/2=3*Cs*Vref/8=0.375*Vref. The negative (b=−1) bit decision yields the same maximal value for Q (the input signal Vin gets the opposite value, and the connection of the
capacitor 102, thecapacitor 104 to the reference voltage Vref are cross-swapped). Since there is no load at all applied on the reference voltage Vref for a null bit decision (b=0), the total variation (with the input signal Vin) of the charge taken from the reference voltage Vref is thus: Q,var=Q,MAX−0=0.375*Cs*Vref. - Since the loading of the reference voltage Vref depends on the input signal Vin, complete settling of voltage on the sampling capacitor is then needed to prevent sampling noise. However, this implies a longer sampling time. If high speed is required, then more power has to be burned on the reference voltage buffer to shorten the settling time.
- Therefore, there is a need to provide a solution capable of reducing the power consumption and eliminating the need of full settling of the voltage on the sampling capacitor charged by a voltage reference.
- To overcome the drawback, the present invention proposes a novel topology of signal independent voltage reference loading for the pipelined ADC.
- One objective of the present invention is to provide a signal independent voltage reference loading circuit to construct a concise and power saving pipelined ADC.
- Another objective of the present invention is to provide a signal independent voltage reference loading circuit to release the speed requirement of the voltage reference buffer.
- Still another objective of the present invention is to provide a signal independent voltage reference loading circuit to construct a 4x interleaved pipelined ADC.
- To achieve the foregoing objectives, the present invention provides a 1-bit cell circuit used in a pipelined analog to digital converter, having a sampling phase including a first half period and a second half period and a charge transfer phase in a cycle. The 1-bit cell circuit comprises a differential amplifier, having a positive input end, a negative input end, a positive output end and a negative output end, for outputting a residue output signal; a first capacitor, having a first plate and a second plate, and a second capacitor, having a third plate and a fourth plate, wherein said first plate is connected to a positive input signal, said third plate is connected to a negative input signal, and said second plate and said fourth plate are both connected to an input common mode reference ground during said sampling phase, and said first plate is connected to said third plate, said second plate is connected to said negative input end of said differential amplifier, and said fourth plate is connected to said positive input end of said differential amplifier during said charge transfer phase; and a third capacitor, having a fifth plate and a sixth plate, and a fourth capacitor, having a seventh plate and an eighth plate, wherein said fifth plate and said seventh plate are connected to an output common mode reference ground, and said sixth plate and said eighth plate are both connected to said input common mode reference ground during said first half period of said sampling phase; said fifth plate is connected to a first reference contact, said seventh plate is connected to a second reference contact, and said sixth plate and said eighth plate are both connected to said input common mode reference ground during said second half period of said sampling phase; and said fifth plate is connected to said positive output end of said differential amplifier, said sixth plate is connected to said negative input end of said differential amplifier, said seventh plate is connected to said negative output end of said differential amplifier and said eighth plate is connected to said positive input end of said differential amplifier during said charge transfer phase; wherein voltages of said first reference contact and said second reference contact are determined according to an input voltage and a reference voltage, and said input voltage is a voltage difference between said positive input signal and said negative input signal.
- In addition, according to another embodiment of the present invention, a 1-bit cell circuit is used in a pipelined analog to digital converter, having a sampling phase including a first half period and a second half period and a charge transfer phase in a cycle. The 1-bit cell circuit comprises a differential amplifier, having a pair of differential input ends and a pair of differential output ends, for outputting a differential residue output signal; a first pair of capacitors, having a pair of first top plates and a pair of first bottom plates, wherein said pair of first top plates are connected to a differential input signal, and said first bottom plates are connected to an input common mode reference ground during said sampling phase, and said first top plates are connected to each other and said first bottom plates are connected to said differential input ends of said differential amplifier during said charge transfer phase; and a second pair of capacitors, having a pair of second top plates and a pair of second bottom plates, wherein said second top plates are connected to an output common mode reference ground and said second bottom plates are connected to an input common mode reference ground during said first half period of said sampling phase; said second top plates are connected to a pair of reference contacts and said second bottom plates are connected to said input common mode reference ground during said second half period of said sampling phase; and said second top plates are connected to said differential output ends of said differential amplifier and said second bottom plates are connected to said differential input ends of said differential amplifier during said charge transfer phase; wherein voltages of said reference contacts are determined according to an input voltage of said differential input signals and a reference voltage.
- In addition, according to another embodiment of the present invention, a 1-bit cell circuit is used in a pipelined analog to digital converter. The 1-bit cell circuit comprises a reference buffer for providing a reference voltage; a sample and charge transfer circuit for receiving an input signal to generate an output signal; and a dump circuit for dumping said reference voltage; wherein said reference buffer selectively connects to one of said sample and charge transfer circuit and said dump circuit according to said input signal.
- To make it easier for our examiner to understand the objective of the invention, its structure, innovative features, and performance, we use a preferred embodiment together with the accompanying drawings for the detailed description of the invention.
-
FIG. 1 is the typical architecture of a prior art 1-bit cell of a pipelined ADC. -
FIG. 2 a is an illustrating diagram of sampling phase of the prior art 1-bit cell. -
FIG. 2 b is an illustrating diagram of charge transfer phase of the prior art 1-bit cell. -
FIG. 3 is the circuit diagram of a 1-bit cell using a signal independent voltage reference loading circuit according to a preferred embodiment of the present invention. -
FIG. 4 a is an illustrating diagram of a first half period of sampling phase of the 1-bit cell inFIG. 3 . -
FIG. 4 b is an illustrating diagram of a second half period of sampling phase of the 1-bit cell inFIG. 3 . -
FIG. 4 c is an illustrating diagram of a charge transfer phase of the 1-bit cell inFIG. 3 . -
FIG. 5 is a circuit diagram of a cascade of two 1-bit cells according to a preferred embodiment of the present invention. - The present invention will be described in more detail hereinafter with reference to the accompanying drawings that show the preferred embodiment of the invention.
- Please refer to
FIG. 3 a˜b, which shows the circuit diagram of a 1-bit cell using a signal independent voltage reference loading circuit according to a preferred embodiment of the present invention. The 1-bit cell ADC procedure includes a first half period of sampling phase, a second half period of sampling phase and a charge transfer phase during a converting period. As shown inFIG. 3 a, the 1-bit cell ADC serve as a sample and charge transfer circuit comprises an OTA (Operational Transconductance Amplifier ) 301, six matchedcapacitors 302˜307, eightsampling phase switches 308˜315, seven chargetransfer phase switches 316˜322, two 323 and 324, twomultiplexers 337 and 338 and acomparators voltage reference buffer 339. As shown inFIG. 3 b, a dump circuit comprises two 330 and 336, two matchedmultiplexers 325 and 331, fourcapacitors sampling phase switches 326˜327 and 332˜333, and four chargetransfer phase switches 328˜329 and 334˜335. - In the 1-bit cell circuit, the
OTA 301, having a positive input end, a negative input end, a positive output end and a negative output end, is used to generate a residue output Vout according to an input signal Vin, wherein the residue output Vout is composed of a positive output Voutp and a negative output Voutn, and the input signal Vin is composed of a positive input signal Vinp and a negative input signal Vinn. The eight matchedcapacitors 302˜307, 325 and 331 are implemented to function as follows. In the circuit, the 302, 303 can be viewed as a first capacitor and thecapacitors 305, 306 can be viewed as a second capacitor. The first capacitor and the second capacitor can be viewed as a first pair of capacitors; thecapacitors 304, 307 can be viewed as a second pair of capacitors; and thecapacitors 325, 331 can be viewed as a third pair of capacitors, wherein the first pair of capacitors are used to hold the sampled voltage of the input signal, and the second pair of capacitors are used to hold the sampled voltage of a differential voltage, which is selected from one of a reference voltage and an output common mode reference ground, and the third pair of capacitors are used to serve as a dump circuit for the reference voltage when the reference voltage is not selected to charge the second pair of capacitors. The first capacitor, composed ofcapacitors 302 and 303, both implemented according to a predetermined capacitance, is used to hold a sampled voltage of the positive input signal Vinp. Thecapacitors capacitor 304, implemented according to the predetermined capacitance, is used to hold one voltage selected from a group consisting of a positive reference voltage Vrefp, a negative reference voltage Vrefn and the output common mode reference ground. The second capacitor, composed of 305 and 306, both implemented according to the predetermined capacitance, is used to hold a sampled voltage of the negative input signal Vinn. Thecapacitors capacitor 307, implemented according to the predetermined capacitance, is used to hold one voltage selected from a group consisting of the positive reference voltage Vrefp, the negative reference voltage Vrefn and the output common mode reference ground. Thecapacitor 325, implemented according to the predetermined capacitance, is used to provide a dump load for the positive reference voltage Vrefp when the output voltage of themultiplexer 323 and the output voltage of themultiplexer 324 both connect to the output common mode reference ground. Thecapacitor 331, implemented according to the predetermined capacitance, is used to provide a dump load for the negative reference voltage Vrefn when the output voltage of themultiplexer 323 and the output voltage of themultiplexer 324 both connect to the output common mode reference ground. - The twelve
sampling phase switches 308˜315, 326˜327 and 332˜333 are implemented to function during the sampling phase as follows. The 308 and 309 are used to sample the positive input signal Vinp into theswitches 302 and 303. Thecapacitors switch 310 is used to connect the bottom plate of thecapacitor 304 to an input common mode reference ground and theswitch 311 is used to connect the top plate of thecapacitor 304 to the output of themultiplexer 323. The 312 and 313 are used to sample the negative input signal Vinn into theswitches 305 and 306. Thecapacitors switch 314 is used to connect the bottom plate of thecapacitor 307 to the input common mode reference ground and theswitch 315 is used to connect the top plate of thecapacitor 307 to the output of themultiplexer 324. Theswitch 326 is used to connect the bottom plate of thecapacitor 325 to the input common mode reference ground and theswitch 327 is used to connect the top plate of thecapacitor 325 to the output of themultiplexer 330. Theswitch 332 is used to connect the bottom plate of thecapacitor 331 to the input common mode reference ground and theswitch 333 is used to connect the top plate of thecapacitor 331 to the output of themultiplexer 336. - The eleven charge
transfer phase switches 316˜322, 328˜329 and 334˜335 are implemented to function during the charge transfer phase as follows. Theswitch 316 is used to interconnect the top plates of the 302, 303, 305 and 306. Thecapacitors switch 317 is used to connect the bottom plates of the 302, 303 to the negative input end of thecapacitors OTA 301. Theswitch 318 is used to connect the bottom plate of thecapacitor 304 to the negative input end of theOTA 301. Theswitch 319 is used to connect the bottom plate of thecapacitors 304 to the positive output end of theOTA 301. Theswitch 320 is used to connect the bottom plates of the 305 and 306 to the positive input end of thecapacitors OTA 301. Theswitch 321 is used to connect the bottom plate of thecapacitor 307 to the positive input end of theOTA 301. Theswitch 322 is used to connect the top plate of thecapacitor 307 to the negative output end of theOTA 301. As a result, thecapacitors 302˜307 and theOTA 301 together form a negative feedback circuit to generate the residue output Vout during the charge transfer phase. Besides, theswitch 328 is used to connect the bottom plate of thecapacitor 325 to the input common mode reference ground and theswitch 329 is used to connect the top plate of thecapacitor 325 to the output common mode reference ground. Theswitch 334 is used to connect the bottom plate of thecapacitor 331 to the input common mode reference ground and theswitch 335 is used to connect the top plate of thecapacitor 331 to the output common mode reference ground. Themultiplexer 323 is used to provide a first multiplexer output voltage selected from a group consisting of the negative reference voltage Vrefn, the positive reference voltage Vrefp and the output common mode reference ground according to two select signals bp&e and bn&e, wherein bp&e is logic-AND of a positive bit signal bp and a clock signal e, and bn&e is logic-AND of a negative bit signal bn and the clock signal e, and the rising edge of the clock e is the breaking point of the first half period and the second half period of sampling phase. When bp&e=0 and bn&e=0, the first multiplexer output voltage is connected to the output common mode reference ground; when bp&e=0 and bn&e=1, the first multiplexer output voltage is connected to the positive reference voltage Vrefp; and when bp&e=1 and bn&e=0, the first multiplexer output voltage is connected to the negative reference voltage Vrefn. Therefore, the first multiplexer output voltage is connected to the output common mode reference ground during the first half period of sampling phase, and connected to one of the negative reference voltage Vrefn, the positive reference voltage Vrefp and the output common mode reference ground according to the bp&e and bn&e during the second half period of sampling phase. - The
multiplexer 324 is used to provide a second multiplexer output voltage selected from a group consisting of the negative reference voltage Vrefn, the positive reference voltage Vrefp and the output common mode reference ground according to bp&e and bn&e. When bp&e=0 and bn&e=0, the second multiplexer output voltage is connected to the output common mode reference ground; when bp&e=0 and bn&e=1, the second multiplexer output voltage is connected to the negative reference voltage Vrefn; and when bp&e=1 and bn&e=0, the second multiplexer output voltage is connected to the positive reference voltage Vrefp. Therefore, the second multiplexer output voltage is connected to the output common mode reference ground during the first half period of sampling phase, and connected to one of the negative reference voltage Vrefn, the positive reference voltage Vrefp and the output common mode reference ground according to the bp&e and bn&e during the second half period of sampling phase. - The
multiplexer 330 is used to provide a third multiplexer output voltage selected from a group consisting of the positive reference voltage Vrefp and the output common mode reference ground according to a select signal e&˜(bp+bn). When e&˜(bp+bn)=0, the third multiplexer output voltage is connected to the output common mode reference ground; when e&˜(bp+bn)=1, the third multiplexer output voltage is connected to the positive reference voltage Vrefp. Therefore, the third multiplexer output voltage is connected to the output common mode reference ground during the first half period of sampling phase, and connected to one of the positive reference voltage Vrefp and the output common mode reference ground according to the e&˜(bp+bn) during the second half period of sampling phase. - The
multiplexer 336 is used to provide a fourth multiplexer output voltage selected from a group consisting of the negative reference voltage Vrefn and the output common mode reference ground according to a select signal e&˜(bp+bn). When e&˜(bp+bn)=0, the fourth multiplexer output voltage is connected to the output common mode reference ground; when e&˜(bp+bn)=1, the fourth multiplexer output voltage is connected to the negative reference voltage Vrefn. Therefore, the fourth multiplexer output voltage is connected to the output common mode reference ground during the first half period of sampling phase, and connected to one of the negative reference voltage Vrefp and the output common mode reference ground according to the e&˜(bp+bn) during the second half period of sampling phase. - The
comparator 337, having a first positive comparing input end, a first negative comparing input end and a first comparing output end, is used to generate the positive bit signal bp according to voltage comparison of the input signal Vin and a first reference voltage Vref/4 at the rising edge of the clock e, wherein the Vref is equal to the voltage difference between the Vrefp and the Vrefn. Thecomparator 338, having a second positive comparing input end, a second negative comparing input end and a second comparing output end, is used to generate the negative bit signal bn according to voltage comparison of the input signal Vin and a second reference voltage −Vref/4 at the rising edge of the clock e, wherein (bp, bn) is (0,0) when Vin is between −Vref/4 and Vref/4; (bp, bn) is (0,1) when Vin is smaller than −Vref/4; and (bp, bn) is (1,0) when Vin is greater than Vref/4. - The
voltage reference buffer 339 is used to provide the Vrefp and the Vrefn, and thevoltage reference buffer 339 can be an OTA or a low output impedance amplifier. - The configurations of the first half period of sampling phase, the second half period of sampling phase and the charge transfer phase of the 1-bit cell circuit according to a preferred embodiment of the present invention is shown in
FIG. 4 a˜c. Please refer toFIG. 4 a, which shows an illustrating diagram of the first half period of sampling phase of the 1-bit cell inFIG. 3 . As shown inFIG. 4 a, the top plates of the 302 and 303 are connected to the positive terminal of the Vin, and the top plates of thecapacitors 305 and 306 are connected to the negative terminal of the Vin, and all the bottom plates of thecapacitors 302, 303, 305 and 306 are connected to the input common mode reference ground. The bottom plates of thecapacitors 304, 307 are connected to the input common mode reference ground, and the top plate of thecapacitors capacitor 304 is connected to a first reference contact, which is the output of themultiplexer 323, being connecting to the output common mode reference ground according to the bp&e and bn&e during the first half period of sampling phase, and the top plate of thecapacitor 307 is connected to a second reference contact, which is the output of themultiplexer 324, being connecting to the output common mode reference ground according to the bp&e and bn&e during the first half period of sampling phase. The bottom plates of the 325, 331 are connected to the input common mode reference ground, and the top plates of thecapacitor 325, 331 are respectively connected to the output ends of thecapacitors 330, 336, being both connecting to the output common mode reference ground according to the e&˜(bp+bn). Themultiplexer comparator 337 generates the positive bit signal bp and thecomparator 338 generates the negative bit signal bn at the end of the first half period of sampling phase. -
FIG. 4 b shows an illustrating diagram of the second half period of sampling phase of the 1-bit cell inFIG. 3 . As shown inFIG. 4 b, the top plates of the 302 and 303 are connected to the positive terminal of the Vin, and the top plates of thecapacitors 305 and 306 are connected to the negative terminal of the Vin, and all the bottom plates of thecapacitors 302, 303, 305 and 306 are connected to the input common mode reference ground. The top plate of thecapacitors capacitor 304 is connected to the first reference contact, which is the output of themultiplexer 323, being connecting to one of the negative reference voltage Vrefn, the positive reference voltage Vrefp and the output common mode reference ground according to the bp&e and bn&e. The top plate of thecapacitor 307 is connected to the second reference contact, which is the output of themultiplexer 324, being connecting to one of the negative reference voltage Vrefn, the positive reference voltage Vrefp and the output common mode reference ground according to the bp&e and bn&e. The voltage difference between the top plates of the 304 and 307 is therefore equal to a first differential voltage (bn-bp)*Vref. The bottom plate of thecapacitor capacitor 325 is connected to the input common mode reference ground and the top plate of thecapacitors 325 is connected to the output of themultiplexer 330, which is connected to one of the positive reference voltage Vrefp and the output common mode reference ground according to the e&˜(bp+bn). The bottom plate of thecapacitor 331 is connected to the input common mode reference ground and the top plate of thecapacitors 331 is connected to the output of themultiplexer 336, which is connected to one of the negative reference voltage Vrefn and the output common mode reference ground according to the e&˜(bp+bn). The voltage difference between the top plates of the 325 and 331 is therefore equal to a second differential voltage (1-bn-bp)*Vref. It can be proven that the sum of the absolute value of the first differential voltage (bn-bp)*Vref and the absolute value of the second differential voltage (1-bn-bp)*Vref is equal to Vref no matter what the (bp,bn) may be.capacitor -
FIG. 4 c shows an illustrating diagram of the charge transfer phase of the 1-bit cell inFIG. 3 . During the charge transfer phase, the top plates of thecapacitors 302˜303, 305˜306 are connected together, and the bottom plates of thecapacitors 302˜303 are connected to the negative input end of theOTA 301, and the bottom plates of thecapacitors 305˜306 are connected to the positive input end of theOTA 301; the bottom plate of thecapacitor 304 is connected to the negative input end of theOTA 301, and the top plate of thecapacitor 304 is connected to the positive output end of theOTA 301, and the bottom plate of thecapacitor 307 is connected to the positive input end of theOTA 301, and the top plate of thecapacitor 307 is connected to the negative output end of theOTA 301; and the bottom plates of the 325, 331 are connected to the input common mode reference ground and the top plates of thecapacitor 325, 331 are connected to the output common mode reference ground. In this phase, the Vref is not used, and due to the virtual ground of thecapacitor OTA 301 in negative feedback, the charge in thecapacitors 302˜307 will be redistributed to generate the residue output Vout=2×Vin−(bp−bn)×Vref. - Please refer to
FIG. 5 , which shows a circuit diagram of a cascade of two 1-bit cells in a pipelined ADC according to a preferred embodiment of the present invention. As shown inFIG. 5 , the residue output of a 1-bit cell 510 is connected to the input of a 1-bit cell 520, and the 1-bit cell 510 generates a positive bit signal bp1 and a negative bit signal bn1, and the 1-bit cell 520 generates a positive bit signal bp2 and a negative bit signal bn2. In an n-bit pipelined converter with resolution n>2, there will be at each clock phase more than one cell using the reference voltage (if n is even, there will be always n/2 cells using the reference voltage; if n is odd, there will be alternately (n+1)/2 and (n−1)/2 cells using the reference voltage). In the second half period of all clock phases, Vref is not used at all, i.e., not used by the cell 510 (and all other odd cells not depicted), nor by the cell 520 (and all other even cells not depicted). This allows the reference voltage to be shared with another set of cells operated with half a clock phase delay, to build a 4× time-interleaved ADC. This feature cannot be achieved in the prior art implementation, not really because the reference voltage is used for the complete duration of each clock period, but rather because of the cascade settling of the reference buffer and OTA in each 1-bit cell. For example, if, at the half instance of a clock phase, a new extra load is suddenly connected to the reference buffer, a resulted glitch will propagate to all the 1-bit cells currently in settling, that will significantly delay their final settling. As a matter of fact, the novel implementation of the present invention uses the reference voltage only for half the duration of each clock phase, while the maximal current (or load) sunk from the reference voltage is staying the same. This fact, together with the relaxed requirement on the reference settling, makes the 4× time-interleaved ADC possible. - In summary, according to the present invention, the positive reference voltage Vrefp has a load capacitor with constant initial charge no matter what bp and bn may be during the second half period of sampling phase, and the load capacitor can be the
capacitor 304 when bp&e=0 and bn&e=1, or be thecapacitor 307 when bp&e=1 and bn&e=0, or be thecapacitor 325 when bp&e=0 and bn&e=0 (i.e., e&˜(bp+bn)=1); and the negative reference voltage Vrefn has a load capacitor with constant initial charge no matter what bp and bn may be during the second half period of sampling phase, and the 32 1, gload capacitor can be thecapacitor 307 when bp&e=0 and bn&e=1, or be thecapacitor 304 when bp&e=1 and bn&e=0, or be thecapacitor 331 when bp&e=0 and bn&e=0 (i.e., e&˜(bp+bn)=1). - As a result, the sum of the absolute value of a first differential voltage, which is the voltage difference between the first multiplexer output voltage and the second multiplexer output voltage, and the absolute value of a second differential voltage, which is the voltage difference between the third multiplexer output voltage and the fourth multiplexer output voltage, is equal to a settling voltage of the reference voltage Vref at the end of the second half period of sampling phase, wherein the reference voltage Vref is the voltage difference between the positive reference voltage Vrefp and the negative reference voltage Vrefn. The
325 and 331 therefore serve as a dump circuit for the positive reference voltage Vrefp and the negative reference voltage Vrefn respectively when thecapacitors 304 and 307 are isolated from the positive reference voltage Vrefp and the negative reference voltage Vrefn by thecapacitors 323 and 324 in the case of (bp, bn)=(0,0).multiplexers - Therefore, a novel 1-bit cell according to the present invention is proposed. The novel 1-bit cell ensures that the amount of charges provided by the reference buffer in each clock period is not dependent on the signal currently converted. Besides, assuming that the clock has achieved the minimal jitter required for aperture noise concerns on high frequency input signal, the novel 1-bit cell ensures both a constant load and a constant settling time for the reference buffer to have a constant settling, even if incomplete, whereas, with a load depending on input signal, full settling is the only way to guaranty a constant final value. Furthermore, it is known that the group delay of a 1-bit cell is critical for the data rate of a pipelined converter in that a shorter group delay will result in a higher data rate, and also critical for the total conversion delay, the time elapsed between signal sampling and full n-bit code delivery, which is of extreme importance in some application like AGC (Automatic Gain Control) or other closed loop systems. The 1-bit cell of the present invention preserves the fundamental 1-clock phase group delay, i.e., a new residue output is delivered to the next cell exactly one clock phase after an old residue was sampled from the previous cell, so it can offer a smallest group delay, a smallest total conversion delay and therefore a high data rate. The present invention thus possesses the following advantages. First, the settling voltage of the second pair of capacitors needs not equal to the reference voltage Vref, so the speed requirement on the reference buffer can be relaxed, and the power needed by the reference buffer is thereby reduced. Second, since during the second half period of sampling phase the voltage reference Vref always has a pair of capacitors implemented according to a predetermined capacitance, and the pair of capacitors being with constant initial charge, the total charge supplied by the voltage reference Vref is therefore constant, i.e. independent of the input signal. Even though there exists mismatch between the capacitors, for example, ˜0.1%, the variation of the charge taken from the voltage reference Vref will be ˜0.1%×the predetermined capacitance×Vref/2, much smaller than the 0.75×Cs×Vref/2 of the prior art. This ensures a constant value to the final output voltage of the reference buffer, and thus avoids any samples interference of the input signal and greatly improves the linearity of the ADC operation. Third, compared to the prior art 1-bit cell circuit, the settling of the reference buffer is not interfering with the setting of the OTA's (no cascade setting), and thereby avoids any slow down of each cell settling. Fourth, since the reference buffer is only used half the time, it can, within same power, be shared by one more ADC time-shifted by half a clock cycle to build a 4× interleaved ADC even if scaling down design is progressively applied from 1st (MSB) to last (LSB) cells for optimal power and area concerns. Finally, the novel design of the 1-bit cell of the present invention offers the signal independent loading of the reference voltage with only 1-clock phase group delay.
- Through the implementation of the 1-bit cell circuit used in pipelined ADCs according to a preferred embodiment of the present invention, a novel pipelined ADC is presented. The topology of the present invention saves much more power than prior art circuits and can be utilized to build a 4× interleaved ADC, so the contribution of the present invention is indeed worthy of the grant of a patent.
- While the invention has been described by way of example and in terms of a preferred embodiment, it is to be understood that the invention is not limited thereto. To the contrary, it is intended to cover various modifications and similar arrangements and procedures, and the scope of the appended claims therefore should be accorded the broadest interpretation so as to encompass all such modifications and similar arrangements and procedures.
- In summation of the above description, the present invention herein enhances the performance than the conventional structure and further complies with the patent application requirements and is submitted to the Patent and Trademark Office for review and granting of the commensurate patent rights.
Claims (15)
Priority Applications (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US12/489,778 US7852254B1 (en) | 2009-06-23 | 2009-06-23 | 1-bit cell circuit used in a pipelined analog to digital converter |
| TW101136840A TWI508459B (en) | 2009-06-23 | 2009-07-31 | 1-bit cell circuit for pipeline analog-to-digital converters |
| TW098125838A TWI384765B (en) | 2009-06-23 | 2009-07-31 | 1-bit cell circuit for pipeline analog to digital converters |
| CN2009101657950A CN101931410B (en) | 2009-06-23 | 2009-08-10 | 1-bit cell circuit used in a pipelined analog to digital converter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US12/489,778 US7852254B1 (en) | 2009-06-23 | 2009-06-23 | 1-bit cell circuit used in a pipelined analog to digital converter |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| US7852254B1 US7852254B1 (en) | 2010-12-14 |
| US20100321220A1 true US20100321220A1 (en) | 2010-12-23 |
Family
ID=43303125
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| US12/489,778 Active 2029-09-03 US7852254B1 (en) | 2009-06-23 | 2009-06-23 | 1-bit cell circuit used in a pipelined analog to digital converter |
Country Status (3)
| Country | Link |
|---|---|
| US (1) | US7852254B1 (en) |
| CN (1) | CN101931410B (en) |
| TW (2) | TWI384765B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10498228B1 (en) * | 2010-12-20 | 2019-12-03 | Anadyne, Inc. | Sub-circuit with cyclically swapped standby circuits |
| US20210376726A1 (en) * | 2020-01-15 | 2021-12-02 | Teng Long | Mixed analog-to-digital converter circuit |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4956573B2 (en) * | 2009-03-19 | 2012-06-20 | 株式会社東芝 | Amplifier circuit and magnetic sensor |
| US8405537B2 (en) * | 2011-08-11 | 2013-03-26 | Pixart Imaging Inc. | Systems, devices and methods for capacitor mismatch error averaging in pipeline analog-to-digital converters |
| US8754800B2 (en) * | 2012-09-29 | 2014-06-17 | Intel Corporation | Methods and arrangements for high-speed analog-to-digital conversion |
| US10547321B2 (en) * | 2017-10-24 | 2020-01-28 | Microchip Technology Incorporated | Method and apparatus for enabling wide input common-mode range in SAR ADCS with no additional active circuitry |
| CN110768669B (en) * | 2018-07-26 | 2023-09-08 | 瑞昱半导体股份有限公司 | Analog-to-digital converter |
| US10461749B1 (en) * | 2018-10-25 | 2019-10-29 | Shenzhen GOODIX Technology Co., Ltd. | Ground intermediation for inter-domain buffer stages |
| US10733391B1 (en) | 2019-03-08 | 2020-08-04 | Analog Devices International Unlimited Company | Switching scheme for low offset switched-capacitor integrators |
| CN110324043B (en) * | 2019-04-24 | 2023-06-30 | 矽力杰半导体技术(杭州)有限公司 | Pseudo-differential analog-to-digital converter |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5764176A (en) * | 1992-11-19 | 1998-06-09 | Vlsi Technology, Inc. | Pipelined analog to digital converters and interstage amplifiers for such converters |
| US6437720B1 (en) * | 2001-02-16 | 2002-08-20 | Conexant Systems, Inc. | Code independent charge transfer scheme for switched-capacitor digital-to-analog converter |
| US6573850B1 (en) * | 2002-05-24 | 2003-06-03 | Wolfson Microelectronics Limited | Digital-to-analogue converter circuits |
| US7148834B2 (en) * | 2003-09-24 | 2006-12-12 | Infineon Technologies Ag | Analog/digital converter and method for operating an analog/digital converter |
| US20090115523A1 (en) * | 2007-11-05 | 2009-05-07 | Panasonic Corporation | Discrete time amplifier circuit and analog-digital converter |
| US7812753B1 (en) * | 2009-03-30 | 2010-10-12 | Dialog Semiconductor Gmbh | Tri-level dynamic element matcher allowing reduced reference loading and DAC element reduction |
| US7821433B2 (en) * | 2008-03-27 | 2010-10-26 | Seiko Epson Corporation | Pipeline-type analog-to-digital converter |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3211793B2 (en) * | 1999-01-08 | 2001-09-25 | 日本電気株式会社 | AD converter |
| US6545624B2 (en) * | 2000-02-11 | 2003-04-08 | Hyundai Electronics Industries Co., Ltd. | Image sensor with analog-to-digital converter that generates a variable slope ramp signal |
| CN1185794C (en) * | 2000-08-31 | 2005-01-19 | 凌阳科技股份有限公司 | Charge Integral Circular Operation Analog-to-Digital Converter and Its Conversion Method |
| US6977601B1 (en) * | 2004-01-29 | 2005-12-20 | Raytheon Company | Low power current input delta-sigma ADC using injection FET reference |
| US7315200B2 (en) * | 2004-03-31 | 2008-01-01 | Silicon Labs Cp, Inc. | Gain control for delta sigma analog-to-digital converter |
| US6956519B1 (en) * | 2004-10-11 | 2005-10-18 | Faraday Technology Corp. | Switched capacitor circuit of a pipeline analog to digital converter and a method for operating the switched capacitor circuit |
| US7268718B1 (en) * | 2006-07-17 | 2007-09-11 | Fortemedia, Inc. | Capacitor-based digital-to-analog converter for low voltage applications |
| US7595748B2 (en) * | 2007-07-23 | 2009-09-29 | Mediatek Inc. | Method of gain error calibration in a pipelined analog-to-digital converter or a cyclic analog-to-digital converter |
-
2009
- 2009-06-23 US US12/489,778 patent/US7852254B1/en active Active
- 2009-07-31 TW TW098125838A patent/TWI384765B/en not_active IP Right Cessation
- 2009-07-31 TW TW101136840A patent/TWI508459B/en not_active IP Right Cessation
- 2009-08-10 CN CN2009101657950A patent/CN101931410B/en not_active Expired - Fee Related
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5764176A (en) * | 1992-11-19 | 1998-06-09 | Vlsi Technology, Inc. | Pipelined analog to digital converters and interstage amplifiers for such converters |
| US6437720B1 (en) * | 2001-02-16 | 2002-08-20 | Conexant Systems, Inc. | Code independent charge transfer scheme for switched-capacitor digital-to-analog converter |
| US6573850B1 (en) * | 2002-05-24 | 2003-06-03 | Wolfson Microelectronics Limited | Digital-to-analogue converter circuits |
| US7148834B2 (en) * | 2003-09-24 | 2006-12-12 | Infineon Technologies Ag | Analog/digital converter and method for operating an analog/digital converter |
| US20090115523A1 (en) * | 2007-11-05 | 2009-05-07 | Panasonic Corporation | Discrete time amplifier circuit and analog-digital converter |
| US7821433B2 (en) * | 2008-03-27 | 2010-10-26 | Seiko Epson Corporation | Pipeline-type analog-to-digital converter |
| US7812753B1 (en) * | 2009-03-30 | 2010-10-12 | Dialog Semiconductor Gmbh | Tri-level dynamic element matcher allowing reduced reference loading and DAC element reduction |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10498228B1 (en) * | 2010-12-20 | 2019-12-03 | Anadyne, Inc. | Sub-circuit with cyclically swapped standby circuits |
| US20210376726A1 (en) * | 2020-01-15 | 2021-12-02 | Teng Long | Mixed analog-to-digital converter circuit |
| US11652414B2 (en) * | 2020-01-15 | 2023-05-16 | Teng Long | Mixed analog-to-digital converter circuit |
Also Published As
| Publication number | Publication date |
|---|---|
| TW201101701A (en) | 2011-01-01 |
| CN101931410A (en) | 2010-12-29 |
| US7852254B1 (en) | 2010-12-14 |
| TW201304427A (en) | 2013-01-16 |
| TWI508459B (en) | 2015-11-11 |
| TWI384765B (en) | 2013-02-01 |
| CN101931410B (en) | 2013-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7852254B1 (en) | 1-bit cell circuit used in a pipelined analog to digital converter | |
| US7515086B2 (en) | Pipelined analog-to-digital converter and method of analog-to-digital conversion | |
| US5710563A (en) | Pipeline analog to digital converter architecture with reduced mismatch error | |
| CN108574487B (en) | successive approximation register analog-to-digital converter | |
| US11303292B2 (en) | Analog to digital converter with current steering stage | |
| US6653966B1 (en) | Subranging analog to digital converter with multi-phase clock timing | |
| US7397409B2 (en) | Multi-bit pipeline analog-to-digital converter having shared amplifier structure | |
| US6879277B1 (en) | Differential pipelined analog to digital converter with successive approximation register subconverter stages | |
| US6914550B2 (en) | Differential pipelined analog to digital converter with successive approximation register subconverter stages using thermometer coding | |
| US7339512B2 (en) | Analog-to-digital converter without track-and-hold | |
| JP2002074976A (en) | Sample hold amplifier circuit, pipeline type AD converter using the same, and pipeline type DA converter | |
| WO1995017782A1 (en) | Analog-to-digital conversion circuit with improved differential linearity | |
| US7965217B2 (en) | Apparatus and method for pipelined analog to digital conversion | |
| US10804920B2 (en) | A/D converter | |
| US7978116B2 (en) | Apparatus and method for pipelined analog to digital conversion | |
| US7450050B2 (en) | Switched-capacitor reset architecture for opamp | |
| US7570192B2 (en) | Charge-domain pipelined analog-to-digital converter | |
| US7224306B2 (en) | Analog-to-digital converter in which settling time of amplifier circuit is reduced | |
| US8274419B2 (en) | Analog-digital converter with pipeline architecture associated with a programmable gain amplifier | |
| Zahrai et al. | A low-power hybrid ADC architecture for high-speed medium-resolution applications | |
| US20110193736A1 (en) | Switched-capacitor pipeline stage | |
| Liu et al. | A fully differential SAR/single-slope ADC for CMOS imager sensor | |
| CN102790619B (en) | Switched Capacitor Circuit and Pipelined Analog-to-Digital Converter | |
| JP4858962B2 (en) | Semiconductor integrated circuit device | |
| KR100976697B1 (en) | Residual Voltage Amplifier and Analog-to-Digital Converter Using the Same |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| AS | Assignment |
Owner name: MSTAR SEMICONDUCTOR, INC., TAIWAN Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:GINETTI, BERNARD;REEL/FRAME:022862/0546 Effective date: 20090616 |
|
| STCF | Information on status: patent grant |
Free format text: PATENTED CASE |
|
| FPAY | Fee payment |
Year of fee payment: 4 |
|
| MAFP | Maintenance fee payment |
Free format text: PAYMENT OF MAINTENANCE FEE, 8TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1552) Year of fee payment: 8 |
|
| AS | Assignment |
Owner name: MEDIATEK INC., TAIWAN Free format text: MERGER;ASSIGNOR:MSTAR SEMICONDUCTOR, INC.;REEL/FRAME:050665/0001 Effective date: 20190124 |
|
| MAFP | Maintenance fee payment |
Free format text: PAYMENT OF MAINTENANCE FEE, 12TH YEAR, LARGE ENTITY (ORIGINAL EVENT CODE: M1553); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY Year of fee payment: 12 |