WO2008091603A1 - Analog error correction for a pipelined charge-domain a/d converter - Google Patents
Analog error correction for a pipelined charge-domain a/d converter Download PDFInfo
- Publication number
- WO2008091603A1 WO2008091603A1 PCT/US2008/000825 US2008000825W WO2008091603A1 WO 2008091603 A1 WO2008091603 A1 WO 2008091603A1 US 2008000825 W US2008000825 W US 2008000825W WO 2008091603 A1 WO2008091603 A1 WO 2008091603A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- charge
- circuit
- voltage
- charge transfer
- domain
- Prior art date
Links
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61K—PREPARATIONS FOR MEDICAL, DENTAL OR TOILETRY PURPOSES
- A61K8/00—Cosmetics or similar toiletry preparations
- A61K8/18—Cosmetics or similar toiletry preparations characterised by the composition
- A61K8/96—Cosmetics or similar toiletry preparations characterised by the composition containing materials, or derivatives thereof of undetermined constitution
- A61K8/97—Cosmetics or similar toiletry preparations characterised by the composition containing materials, or derivatives thereof of undetermined constitution from algae, fungi, lichens or plants; from derivatives thereof
- A61K8/9783—Angiosperms [Magnoliophyta]
- A61K8/9789—Magnoliopsida [dicotyledons]
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M1/00—Analogue/digital conversion; Digital/analogue conversion
- H03M1/06—Continuously compensating for, or preventing, undesired influence of physical parameters
- H03M1/0602—Continuously compensating for, or preventing, undesired influence of physical parameters of deviations from the desired transfer characteristic
- H03M1/0604—Continuously compensating for, or preventing, undesired influence of physical parameters of deviations from the desired transfer characteristic at one point, i.e. by adjusting a single reference value, e.g. bias or gain error
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61Q—SPECIFIC USE OF COSMETICS OR SIMILAR TOILETRY PREPARATIONS
- A61Q11/00—Preparations for care of the teeth, of the oral cavity or of dentures; Dentifrices, e.g. toothpastes; Mouth rinses
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11C—STATIC STORES
- G11C27/00—Electric analogue stores, e.g. for storing instantaneous values
- G11C27/02—Sample-and-hold arrangements
- G11C27/024—Sample-and-hold arrangements using a capacitive memory element
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61K—PREPARATIONS FOR MEDICAL, DENTAL OR TOILETRY PURPOSES
- A61K2800/00—Properties of cosmetic compositions or active ingredients thereof or formulation aids used therein and process related aspects
- A61K2800/20—Chemical, physico-chemical or functional or structural properties of the composition as a whole
- A61K2800/22—Gas releasing
- A61K2800/222—Effervescent
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61K—PREPARATIONS FOR MEDICAL, DENTAL OR TOILETRY PURPOSES
- A61K2800/00—Properties of cosmetic compositions or active ingredients thereof or formulation aids used therein and process related aspects
- A61K2800/20—Chemical, physico-chemical or functional or structural properties of the composition as a whole
- A61K2800/28—Rubbing or scrubbing compositions; Peeling or abrasive compositions; Containing exfoliants
-
- 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
Definitions
- signals are represented as charge packets. These charge packets are stored, transferred from one storage location to another, and otherwise processed to carry out specific signal-processing functions. Charge packets are capable of representing analog quantities, with the charge-packet size in coulombs being proportional to the signal represented. Charge-domain operations such as charge-transfer are driven by 'clock' voltages, providing discrete-time processing. Thus, charge-domain circuits provide analog, discrete-time signal-processing capability. This capability is well-suited to performing analog-to-digital conversion using pipeline alogorithms.
- Charge-domain circuits are implemented as charge-coupled devices (CCDs), as MOS bucket-brigade devices (BBDs), and as bipolar BBDs.
- CCDs charge-coupled devices
- BBDs MOS bucket-brigade devices
- the present invention pertains to MOS BBDs.
- Pipelined analog-to-digital converters are well-known in the general field of ADC design. They are widely used in applications in which high sample rates and high resolution must be combined. Pipelined ADCs implement the well- known successive-approximation analog-to-digital (A/D) conversion algorithm, in which progressively-refined estimates of an input signal are made at sequential times. In pipelined versions of this algorithm, one or several bits are resolved at each pipeline stage, the quantized estimate is subtracted from the signal, and the residue is propagated to the next pipeline stage for further processing.
- Pipelined ADCs have been implemented using a variety of circuit techniques, including switched-capacitor circuits and charge-domain circuits. The present invention pertains to charge-domain pipelined ADCs employing MOS BBDs.
- the gain between pipeline stages is nominally unity: that is, all net charge present in each stage ideally is transferred to the next stage.
- the charge-transfer gain is less than unity, resulting in errors in the AfD conversion process.
- mismatch of capacitors and of capacitor ratios causes such errors.
- the present invention corrects for errors in BBD-based pipelined ADCs due to both capacitor mismatch and to sub-unity charge-transfer gain. Circuits that implement this correction are compact and temperature-stable, and consume low power.
- a pipelined charge domain circuit using bucket brigade charge transfer comprises a first charge transfer circuit; a second charge transfer circuit; and a node coupled to the first charge transfer circuit and the second charge transfer circuit.
- a clocked capacitor is coupled to the node and to a clocked voltage.
- a conditionally-switched capacitor is also coupled to the node, with the conditionally-switched capacitor driven by a transition voltage.
- An adjustment circuit is provided for adjusting the transition voltage according to conditions detected within the pipelined charge domain circuit.
- Figure 1 shows a simplified circuit diagram of a BBD charge-pipeline stage.
- Figure 2 illustrates voltage waveforms associated with Figure 1.
- Figure 3 shows a BBD charge-pipeline stage including conditional charge addition.
- Figure 4 illustrates voltage waveforms associated with Figure 3.
- Figure 5 shows a BBD charge-pipeline stage including conditional charge addition, with the added charge composed of two independent components.
- Figure 6 shows a two-stage pipeline segment composed of stages like that of ' Figure 3.
- Figure 7 shows a circuit diagram of a capacitor driver with an adjustable voltage transition.
- Figure 8 illustrates voltage waveforms associated with Figure 7.
- Figure 9 shows a circuit diagram of a replica-based circuit for determining a charge-transfer voltage-feedback coefficient.
- Figure 10 illustrates voltage waveforms associated with Figure 9.
- Figure 1 1 shows the adjustment circuit connected to a pipeline stage.
- Figure 12 is a more detailed view of a differential BBD-charge pipeline stage. DETAILED DESCRIPTION OF THE INVENTION
- Voltage Va is a digital clock signal which controls the timing of charge processing in the stage. Other digital clock signals, not shown, may be used to control the activity of the charge-transfer circuits.
- C is comprised of C 5 , the capacitance of capacitor 5, plus any parasitic capacitance of node 2; such parasitic capacitance is usually small and is neglected in this discussion.
- io Charge transfer out of the stage begins at time t 3 when clock voltage Va switches to a low state.
- Capacitor 5 couples this voltage transition to node 2, driving V 2 low as well.
- Charge-transfer circuit 3 absorbs charge from capacitor 5, limiting the negative excursion of node 2, and eventually causing node 2 to settle to voltage 23 at t 4 .
- Voltage 23 is a characteristic of charge-transfer circuit 3, and is i 5 independent of the amount of charge which had been stored on node 2.
- Charge- transfer circuit 3 transfers the charge absorbed from capacitor 5 to node 4 which is part of the stage following the one shown. After t 4 charge transfer is complete.
- charge is transferred into the stage shown in Figure 1 during ti- t 2 ; between times t 2 and t 3 it is temporarily stored on capacitor 5, and is manifested as the value of V 2 ; during times t 3 -t 4 this charge is completely transferred to the next stage; at t 5 the stage returns to its initial state, ready again to receive incoming charge.
- the basic stage shown acts as a shift register for analog charge packets.
- an ADC which implements pipelined successive-approximation ADC algorithms using these operations.
- the present invention provides a way to improve the accuracy of charge-transfer and conditional-charge addition in such an ADC.
- the charge-comparison aspects of ADC implementation are not important, and are not discussed further.
- Conditional charge addition is essential for such understanding, however, and is explained with reference to Figures 3 and 4 below.
- Figure 3 The pipeline stage shown in Figure 3 retains all the elements shown in Figure 1.
- Figure 3 includes two new elements: capacitor 6 (with value C 6 ) connected between charge-storage node 2 and voltage V QRI ; and switch 7 connected between node 2 and voltage Vp. Switch 7 is controlled by a periodic digital clock signal (identified as S 7 in Figure 4).
- FIG 4 shows the operating waveforms of the circuit of Figure 3.
- the initial conditions in Figure 4 are similar to those in Figure 2: Vci is at high voltage45 and V 2 , the voltage of node 2, is at high voltage 41.
- V QR1 is at high voltage 47, and switch 7 is in an off state, indicated by the low value of its control signal S 7 in Figure 4.
- charge is transferred into the stage between ti and t 2 , causing V 2 to fall in proportion to the incoming charge, settling at voltage 42.
- V QRI conditionally switches from its high state 47 to low state 48. This conditional transition of V QRI is coupled via C 6 to node 2 where, because of capacitive division, it produces a similar but smaller voltage change. The voltage at node 2 changes to voltage 49 (dashed line) if V QRI switches, and remains at voltage 42 (solid line) if it does not.
- Vci switches from high voltage 45 to low voltage 46, instigating charge transfer out of the stage.
- node 2 is driven to a lower voltage due to coupling via capacitor 5.
- Charge-transfer circuit 3 removes charge from node 2 and transfers it to the next stage.
- V 2 settles to voltage 43 which is independent of the charge previously on node 2, and charge transfer out of the stage is complete.
- Vci and VQRI return to their initial high states (voltages 45 and 47 respectively). This transition is identical for Va in every clock cycle.
- VQRI may already be at its high voltage 47, depending on whether or not it switched at : 3 A.
- the added switch 7 in Figure 3 is used to restore the voltage (and charge) on node 2 to a repeatable state regardless of the state of VQRI at t 5 .
- voltage 44 Vp
- Q CONST is nominally a fixed charge, since voltages 41, 43, 45, and 46 are all ideally constant. Departures from this ideal case, which constitute one source of charge-transfer imperfection, will be discussed below. As is apparent in Figure 4, ⁇ V QRI is equal to (voltage 48 - voltage 47) if V QRI switches, and is equal to zero if it does not. Note that both charge quantities C O ⁇ V QRI and Q C O NST can be made either positive or negative by appropriate choices of the various voltages.
- the quantity (voltage 48 — voltage 47) is made equal to a reference voltage; for convenience it will be called V RJ .
- the quantity C 6 V RI becomes a reference charge, since C 6 is fixed in a given instantiation.
- the circuit of Figure 3 thus provides one of the two operations required for a charge- domain ADC implementation.
- V QRI can occur at any time between t 0 and t 3 with no change in circuit performance; under some practicable conditions it may also occur in the t 3 - U interval.
- FIG. 5 An example of such a stage is shown in Figure 5.
- This circuit includes, in addition to the elements in Figure 3, additional capacitor 6 A and voltage source V QR2 .
- the operation of such a stage is identical to that of Figure 3, except that at t 3A each of the voltages VQ RI and VQ R2 undergoes an independent conditional transition, of size V R1 and V R2 (or zero) respectively.
- the resulting charge transfer function of the stage is given by:
- QCONST2 (C 5 +C 6 +C 6A )(voltage 41-voltage 43) + C 5 (voltage 46 - voltage 45)
- Equations 1 and IA will be used as the basis for the following discussion.
- the principles described are equally applicable to circuits with more than one conditionally-switched capacitor, as in Figure 5.
- V 43 V 43N + kQou ⁇
- V 43 the actual value of voltage 43
- V 23N the nominal value
- k a coefficient embodying the linear dependence on Q OUT -
- a pipelined charge-domain ADC is composed of multiple stages like that of Figure 3, in which reference charges are conditionally added and subsequently transferred down the pipeline. (As mentioned above, some architectures employ multiple conditional charges per stage.) Equation 6 expresses the output charge of each such stage in terms of that stage's input charge and its conditional capacitor switching. At any stage of the pipeline, the input charge is the sum of the signal- charge input to the pipeline and the cumulative changes due to up-stream pipeline stages according to Equation 6.
- the two-stage pipeline segment shown in Figure 6, which is composed of stages like that in Figure 3.
- the two stages, 61 and 62 each consist of a storage node (67 and 68 respectively), a charge-transfer circuit (65 and 66 respectively), a conditionally-switched capacitor (63 and 64 respectively), a clocked capacitor (601 and 602), and a precharge switch Vp.
- the input charge to the pipeline segment, QPIN is supplied to storage node 67 of stage 61 as indicated.
- the output of this pipeline segment (which is the output charge of stage 62) appears at node 69.
- the conditionally-switched voltages driving capacitors 63 and 64 are V Q RI and VQR 2 respectively. They can each independently be switched, with a step-size of 0 or VRI and 0 or VR 2 respectively, where V R] and VR 2 are reference voltages.
- Such a pipeline operates in two-phases: alternating stages operate on alternating half-cycles of the clock.
- charge is transfered into stage 61 and out of stage 62 on the first half-cycle, and is transferred out of stage 61 and into stage 62 on the second half-cycle. Details of this clocking method are not germane to the subject of the present invention, and are not treated further.
- the input charge to stage 62 in Figure 6 is the output charge from stage 61.
- the output charge from stage 62 can be derived by applying Equation 6 twice:
- Qou ⁇ 62 QiN62(l- ⁇ ) + C 64N ⁇ V QR2 (l- ⁇ ) + Qcc(l- ⁇ ) + C 64E ⁇ V QR2
- Equation 8 shows the cumulative effect of charge -transfer gain and capacitor errors in a two-stage pipeline. The same analysis can be extended to multiple stages and to multiple conditionally-switched capacitors per stage.
- conditionally-added charges from each stage appear at the end of the ADC pipeline with a specific ratio.
- the (non-zero) values of the conditionally-added charges are nominally C 6 3NVRI and C 64 NVR 2 in the first and second stages respectively.
- the conditionally-added charge values, as they appear at the pipeline output are:
- Equation 9 is simplified to:
- Equation 9 The effects of non-zero gain error ⁇ and capacitor errors such as C 63E are evident when Equation 9 is compared to Equation 10.
- One aspect of the present invention provides a way to satisfy Equation 9 when these errors have non-zero values. This can consist of providing an adjustment to the reference voltages V RI and V R2 .
- the nominal capacitance values C 63N and C 64N and the reference voltages V RI and V R2 are chosen to satisfy Equation 5 10; then independently adjustable voltages V A i and V A2 are added to V RI and V R2 .
- the added voltages are now adjusted to force the ratio of the adjusted bracketed terms to equal K. For example, setting: 0
- VAI [ ⁇ - (C 63E /C 63N )]V R i Equation 1 IA and
- VA 2 - (C 64E /C 64N )VR2 Equation 1 1 B 5 results in the desired ratio.
- V A2 is adjusted to correct for the error of capacitor 64 and V AI is adjusted to compensate both for the error of capacitor 63 and for the charge-transfer gain ⁇ .
- V A i is adjusted to correct for the error of capacitor 63 and V A2 is adjusted to compensate both for the error of capacitor 63 and for the charge-transfer gain ⁇ .
- Both solutions 1 1 AJB and 12A/B are useful. Any linear combination of these solutions can be used with the same result.
- This same adjustment principle can be applied in a pipeline with any number of stages. It can also be applied in an ADC design with more than one conditionally-added charge per stage (as in the example of Figure 5). In these cases, a separate adjustment voltage (V A ) is applied to the reference voltage for each conditionally-switched capacitor.
- V A the adjustment voltage
- each such adjustment corrects for the combination of that individual capacitor's error and the total charge-transfer errors of all previous stages.
- the charge-transfer error is not identical from stage to stage, but this adjustment method intrinsically accommodates such variation.
- the same adjustment principle can be applied to a differential pipeline stage; in such an instance it may be preferable to generate a single adjustment voltage V A that is shared between the two members of the differential circuit.
- Figure 1 1 shows an example of a pipeline stage like that of Figure 1 with such an adjustment voltage added.
- Pipeline elements 1, 2, 3, 5, and 6 are equivalent to those in Figure 1.
- the added elements are a driver 8 for conditionally-switched capacitor 6, and an adjustable voltage source 9.
- the output of driver 8 switches between supplied voltage V 8 and the voltage at node 10 in response to a digital control signal SQRI , thus providing the required conditional voltage transition.
- VH and voltage 10 provide the voltages identified as 47 and 48 respectively in
- Adjustable voltage source 9 adjusts the low voltage supplied to capacitor 6 as described above.
- Figure 12 shows a differential pipeline with similar function.
- Each member of the differential pair of pipeline nodes, 2 and 122, is provided with a conditionally- switched capacitor, 6 and 126 respectively.
- the voltage transitions driving capacitors 6 and 126 are provided by drivers 8 and 128, switching between VH and the voltages of nodes 10 and 130, in response to digital control signals SQRIA and SQRIB, respectively.
- the voltages of nodes 10 and 130 are supplied by adjustable voltage sources 9 and 129 respectively.
- Nodes 10 and 130 may be connected to form a single node, and supplied by a single adjustable source such as 9.
- Figure 7 shows a CMOS inverter consisting of PFET 71 and NFET 72, supplied by fixed voltages V H and VL.
- the NFET is connected to VL via resistor 73 at node 75.
- An adjustable current 74 is also connected to node 75.
- the inverter is driven by a logic signal S QR I , and its output constitutes the voltage signal V QRI shown in Figure 3.
- Figure 8 shows operating waveforms for the circuit of Figure 7. At time t 0 logic signal S QRI is at a low state, turning PFET 71 on and NFET 72 off. Voltage
- V QRI is at an initial high value which is equal to V H -
- SQ R I switches to a high logic state, turning FET 71 off and FET 72 on.
- FET 72 thus connects the output V QRI to node 75.
- the adjustable current 74 is set to zero, then the series combination of FET 72 and resistor 73 charges V QRI towards V L , as indicated by voltage curve 81 in Figure 8. Since the load on this circuit consists only of a capacitor (capacitor 6 in Figure 3), there is no DC current through FET 72 or resistor 73, so VQ R I eventually settles to V L . Thus the voltage transition of VQ R I is V L -V H . This quantity constitutes the unadjusted reference voltage V R .
- Adjustable current source 74 which is easily realizable in a practical circuit, thus provides for adjusting the size of the transition in V QRI , as required.
- the required V A values have two components: one which corrects for a capacitor error and one which corrects for charge-transfer gain error.
- the current source 74 can be made to consist of two independent sources in parallel, each controlled independently to correct for one of the errors. The combined currents sum to develop the composite V A value.
- the capacitor-error component corrected by V A can be expected to be temperature-invariant because it is due primarily to geometric variation between capacitors, which occurs during circuit fabrication but does not generally change thereafter.
- an adjustment voltage V A which tracks V R provides a temperature- stable adjustment. Creating a component of V A which tracks V R over temperature using circuits similar to Figure 7 is straightforward in a conventional CMOS process.
- the value of this adjustment voltage can be set by a calibration process carried out during manufacturing test or upon powering-up the circuit, for example.
- the second V A component corrects for charge-transfer gain error.
- This error depends on details of the charge -transfer circuits employed, and depends in general on both fabrication-process variation and on operating temperature.
- Known BBD charge-transfer circuits include both conventional (passive) ones and ones employing active circuitry, such as those described in a previous patent application by the same inventor entitled "Boosted Charge-Transfer Pipeline", U.S. Patent Application No. 11/807,914, filed May 30, 2007, which is hereby incorporated by reference in its entirety. These charge-transfer circuits exhibit both dynamic and static components of charge-transfer gain error.
- One aspect of the present invention provides for generating an adjustment- voltage component which tracks the static component of charge-transfer error.
- the charge-transfer gain error was formulated in terms of the variation of charge-transfer-circuit input voltage with the amount of charge transferred.
- V 43 V 43N + kQou ⁇
- V 23 N the nominal value
- k a coefficient embodying the.linear- dependence on Qoy ⁇ - This formulation encompasses both dyanmic dependence of V 43 on Q O U T (due to incomplete settling) and static dependence.
- the coefficient k reflects only static dependence.
- the charge-transfer gain error ⁇ depends on the voltage-feedback coefficient ⁇ of the charge-transfer circuit and a ratio of pipeline node capacitances. Since the pipeline node-capacitance ratios are known by design within small tolerances (discussed above), the gain error value ⁇ can be derived from a determination of the voltage-feedback coefficient ⁇ .
- Figure 9 shows a circuit which senses the voltage-feedback coefficient ⁇ . It consists of a charge-transfer circuit 91 with input node 92 and output node 93, a voltage source 95 connected to node 93, and a current source 94 connected to node 92.
- Charge-transfer circuit 91 is a replica of the charge-transfer circuits employed in the actual charge pipeline.
- Current source 94 is configured to sink from node 92 a small current which is typical of current levels near the end of the charge-transfer process described above.
- Voltage source 95 provides a voltage at node 93 which is in the range normally occurring at the output of the charge-transfer circuit at the end of the charge transfer process (such as voltage 42 in Figure 4).
- the charge- transfer circuit in Figure 9 is biased in a static condition essentially like its instantaneous condition near the end of the normal (clocked) charge-transfer process.
- FIG. 10 shows the voltages of the two nodes in Figure 9.
- V 93 the voltage of node 93, is driven cyclically by voltage source 95 between two levels 101 and 102 differing by ⁇ V 93 .
- the input voltage of the charge-transfer circuit i.e., the voltage V 92 at node 92
- V 92 at node 92 the voltage V 92 at node 92
- ⁇ V 92 provides a direct measure of ⁇ .
- the voltage change ⁇ V 92 can be converted to a DC voltage using phase-sensitive detection, since the alternating drive voltage V 93 is available as a reference.
- the frequency of this alternating voltage is not critical, and need not be as high as the sample rate of the charge pipeline, as the parameter being sensed changes only slowly (primarily due to chip temperature changes). Because of the low current at node 92, settling of V 92 in response to each V 93 transition is relatively slow, so the operating frequency of this circuit must be limited in order to obtain a valid settled value for ⁇ V 92 .
- two circuits like that of Figure 9, with differing DC voltages supplied by the output-node voltage sources, thus producing two different input- node voltages, can be used to directly generate a static (DC) value of ⁇ V 92 .
- the alternating-voltage method described above is generally more accurate, however, since ⁇ V 92 is quite small (typically only a few mV or tens of mV).
- the charge-transfer circuits may not be of identical design at all stages.
- the ⁇ -sensing circuitry just described consumes very little power, and can practically be reproduced for each charge-transfer circuit design employed.
- the circuitry described provides a correction voltage based on a charge- transfer circuit which is a replica of the charge transfer circuits in the pipeline.
- a replica-based method provides very good tracking over operating conditions, but typically has small initial mismatches. Such initial mismatches can be removed in a calibration operation, at manufacturing test or during circuit power-up for example. After the calibration, the replica-based circuit provides tracking of subsequent changes in operating conditions, including temperature and supply voltage.
- Such an initial calibration step also (simultaneously) provides correction for any dynamic component of charge-transfer gain error which is present under the calibration conditions. Any change of this dynamic error with operating-condition changes (especially temperature) after calibration, however, is not corrected by the techniques of this invention.
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Veterinary Medicine (AREA)
- Theoretical Computer Science (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Botany (AREA)
- Biotechnology (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Microbiology (AREA)
- Mycology (AREA)
- Birds (AREA)
- Epidemiology (AREA)
- Analogue/Digital Conversion (AREA)
- Cosmetics (AREA)
Abstract
A technique for correcting errors in Bucket Brigade Device (BBD)-based pipelined devices, such as Analog-to Digital Converters (ADCs). The gain between pipeline stages is desired to be a specific amount, such as unity: that is, all net charge present in each stage ideally is transferred to the next stage. In practical BBD-based circuits, however, the charge-transfer gain is less than ideal, resulting in errors. The approach described herein provides analog correction of such errors due to both capacitor mismatch and to sub-unity charge-transfer gain. In certain embodiments the adjustment circuit may use an adjustable current source and Field Effect Transistor to introduce the correction. In still other embodiments, the adjustment circuit may determine a voltage-feedback coefficient.
Description
ANALOG ERROR CORRECTION FOR A PIPELINED CHARGE-DOMAIN A/D CONVERTER
RELATED APPLICATION(S) This application claims the benefit of U.S. Provisional Application No.
60/881,967, filed on January 23, 2007. The entire teachings of the above application(s) are incorporated herein by reference.
BACKGROUND OF THE INVENTION
In charge-domain signal-processing circuits, signals are represented as charge packets. These charge packets are stored, transferred from one storage location to another, and otherwise processed to carry out specific signal-processing functions. Charge packets are capable of representing analog quantities, with the charge-packet size in coulombs being proportional to the signal represented. Charge-domain operations such as charge-transfer are driven by 'clock' voltages, providing discrete-time processing. Thus, charge-domain circuits provide analog, discrete-time signal-processing capability. This capability is well-suited to performing analog-to-digital conversion using pipeline alogorithms.
Charge-domain circuits are implemented as charge-coupled devices (CCDs), as MOS bucket-brigade devices (BBDs), and as bipolar BBDs. The present invention pertains to MOS BBDs.
Pipelined analog-to-digital converters (ADCs) are well-known in the general field of ADC design. They are widely used in applications in which high sample rates and high resolution must be combined. Pipelined ADCs implement the well- known successive-approximation analog-to-digital (A/D) conversion algorithm, in which progressively-refined estimates of an input signal are made at sequential times. In pipelined versions of this algorithm, one or several bits are resolved at
each pipeline stage, the quantized estimate is subtracted from the signal, and the residue is propagated to the next pipeline stage for further processing. Pipelined ADCs have been implemented using a variety of circuit techniques, including switched-capacitor circuits and charge-domain circuits. The present invention pertains to charge-domain pipelined ADCs employing MOS BBDs.
SUMMARY OF THE INVENTION
In BBD-based pipelined ADCs, the gain between pipeline stages is nominally unity: that is, all net charge present in each stage ideally is transferred to the next stage. In practical BBD-based circuits, however, the charge-transfer gain is less than unity, resulting in errors in the AfD conversion process. Moreover, in all pipelined ADCs including those employing BBDs, mismatch of capacitors and of capacitor ratios causes such errors.
The present invention corrects for errors in BBD-based pipelined ADCs due to both capacitor mismatch and to sub-unity charge-transfer gain. Circuits that implement this correction are compact and temperature-stable, and consume low power.
In a preferred embodiment, a pipelined charge domain circuit using bucket brigade charge transfer comprises a first charge transfer circuit; a second charge transfer circuit; and a node coupled to the first charge transfer circuit and the second charge transfer circuit. A clocked capacitor is coupled to the node and to a clocked voltage. Furthermore, a conditionally-switched capacitor is also coupled to the node, with the conditionally-switched capacitor driven by a transition voltage. An adjustment circuit is provided for adjusting the transition voltage according to conditions detected within the pipelined charge domain circuit.
BRIEF DESCRIPTION OF THE DRAWINGS
The foregoing will be apparent from the following more particular description of example embodiments of the invention, as illustrated in the accompanying drawings in which like reference characters refer to the same parts throughout the different views. The drawings are not necessarily to scale, emphasis instead being placed upon illustrating embodiments of the present invention.
Figure 1 shows a simplified circuit diagram of a BBD charge-pipeline stage. Figure 2 illustrates voltage waveforms associated with Figure 1.
Figure 3 shows a BBD charge-pipeline stage including conditional charge addition.
Figure 4 illustrates voltage waveforms associated with Figure 3. Figure 5 shows a BBD charge-pipeline stage including conditional charge addition, with the added charge composed of two independent components.
Figure 6 shows a two-stage pipeline segment composed of stages like that of 'Figure 3.
Figure 7 shows a circuit diagram of a capacitor driver with an adjustable voltage transition. Figure 8 illustrates voltage waveforms associated with Figure 7.
Figure 9 shows a circuit diagram of a replica-based circuit for determining a charge-transfer voltage-feedback coefficient.
Figure 10 illustrates voltage waveforms associated with Figure 9. Figure 1 1 shows the adjustment circuit connected to a pipeline stage. Figure 12 is a more detailed view of a differential BBD-charge pipeline stage.
DETAILED DESCRIPTION OF THE INVENTION
A description of example embodiments of the invention follows.
The teachings of all patents, published applications and references cited herein are incorporated by reference in their entirety.
In the following description, all circuits are discussed assuming electrons as the signal-charge carriers and NFETs for signal-charge transfer. Functionally equivalent circuits can be applied equally well using holes as charge carriers, by employing PFETs and reversed signal and control voltage polarities. In the discussion and figures, charge-transfer circuits are represented abstractly and the relevant behavioral aspects of these circuits are described, but in some instances, specific details of the operation of such circuits are understood by those of skill in the art and/or are not relevant to the invention claimed herein, and thus are not provided. The issue of sub-unity charge transfer gain is common to all known charge -transfer circuits.
The basic principle of a BBD pipeline of the general type employed in a preferred embodiment of this invention is described with the aid of Figure 1, which depicts a single stage of such a pipeline. In this stage charge is stored on capacitor
5, which is connected between storage node 2 and voltage Va. Charge enters the stage via charge-transfer circuit 1, and later exits the stage via charge-transfer circuit 3. Voltage Va is a digital clock signal which controls the timing of charge processing in the stage. Other digital clock signals, not shown, may be used to control the activity of the charge-transfer circuits.
Operating waveforms of the pipeline stage are shown in Figure 2. At time t0 clock voltage Va has a positive value 25. V2, the voltage of storage-node 2 in Figure 1, is also at a high initial voltage 21. At ti negative charge begins to be transferred from the previous stage (to the left of Figure 1) via charge-transfer circuit
1 into the stage shown. As this negative charge accumulates on capacitor 5, V2 falls
to a more negative value. The voltage of node 2 settles to a relatively high value 22A if a relatively small negative charge was transferred; with a larger charge transferred, node 2 settles to a more negative voltage 22B. At time t2 charge transfer into the stage is complete. The voltage of node 2 is related to the charge by the well- 5 known expression Q = CV, where C is the total capacitance of node 2. In Figure 1,
C is comprised of C5, the capacitance of capacitor 5, plus any parasitic capacitance of node 2; such parasitic capacitance is usually small and is neglected in this discussion.
io Charge transfer out of the stage begins at time t3 when clock voltage Va switches to a low state. Capacitor 5 couples this voltage transition to node 2, driving V2 low as well. Charge-transfer circuit 3 absorbs charge from capacitor 5, limiting the negative excursion of node 2, and eventually causing node 2 to settle to voltage 23 at t4. Voltage 23 is a characteristic of charge-transfer circuit 3, and is i5 independent of the amount of charge which had been stored on node 2. Charge- transfer circuit 3 transfers the charge absorbed from capacitor 5 to node 4 which is part of the stage following the one shown. After t4 charge transfer is complete.
Finally, at time t5, clock voltage Va returns to its initial state (voltage 25).0 Its positive-going transition is coupled to node 2 by capacitor 5, raising node 2 to voltage 24. Neglecting parasitic capacitance, no charge flows onto or off of node 2 during this transition; the voltage change of V2 is therefore equal to the voltage change of Va during the transition at t5. Since V2's value at the start of this transition, voltge 23, is independent of charge processed, voltage 24 is likewise5 independent of charge processed. This transition completes the operating cycle; the resulting voltage 24 at node 2 is thus the initial voltage for the next cycle. Thus the initial voltage state of the stage is constant- cycle-to-cycle, and voltage 21 = voltage 24. Consequently the initial and final charge on node 2 are also equal, and the charge transferred out is equal to the charge transferred in. 0
In summary: charge is transferred into the stage shown in Figure 1 during ti- t2; between times t2 and t3 it is temporarily stored on capacitor 5, and is manifested
as the value of V2; during times t3-t4 this charge is completely transferred to the next stage; at t5 the stage returns to its initial state, ready again to receive incoming charge. Thus the basic stage shown acts as a shift register for analog charge packets.
The foregoing description is somewhat idealized; it should be understood that practical circuits depart in many details from this description. Such departures include non-zero parasitic capacitance and imperfect charge transfer, for example. These effects, however, do not change the basic operating principles described above. Certain details of circuit operation, such as clocking of the charge-transfer circuits, are also omitted, as they are not pertinent to the present invention.
In order to form a charge-domain ADC from a pipeline composed of stages similar to Figure 1 , a minimum of two operations in addition to charge storage and shifting are required: charges must be compared to a reference value, typically another charge; and a reference charge must be conditionally added to the signal charge (this 'addition' may be of either sign). In a previous patent application by the same inventor entitled "Charge-Domain Pipelined Analog-to-Digital Converter" filed on January 18, 2008 and given serial number 11/ , Attorney Docket
Number 3575.1028-002, which itself claims priority an earlier filed provisional application of the same title filed on January 19, 2007 and given provisional application serial number 60/881,392, both of which are hereby incorporated by reference in their entirety, an ADC is disclosed which implements pipelined successive-approximation ADC algorithms using these operations. The present invention provides a way to improve the accuracy of charge-transfer and conditional-charge addition in such an ADC. For purposes of understanding the present invention, the charge-comparison aspects of ADC implementation are not important, and are not discussed further. Conditional charge addition is essential for such understanding, however, and is explained with reference to Figures 3 and 4 below.
The basic principle employed for conditional charge addition is depicted in Figure 3, with operating waveforms shown in Figure 4. For the purposes of this
discussion, a single-ended stage is shown. In practical ADC designs, differential operation is usually preferred; the present invention is applicable to both single- ended and differential pipeline circuits. More details of the implementation of a differential pipeline stage are shown and described in the aforementioned "Charge- Domain Pipelined Analog-to-Digital Converter" patent application.
The pipeline stage shown in Figure 3 retains all the elements shown in Figure 1. In addition, Figure 3 includes two new elements: capacitor 6 (with value C6) connected between charge-storage node 2 and voltage VQRI ; and switch 7 connected between node 2 and voltage Vp. Switch 7 is controlled by a periodic digital clock signal (identified as S7 in Figure 4).
Figure 4 shows the operating waveforms of the circuit of Figure 3. The initial conditions in Figure 4 are similar to those in Figure 2: Vci is at high voltage45 and V2, the voltage of node 2, is at high voltage 41. In addition, VQR1 is at high voltage 47, and switch 7 is in an off state, indicated by the low value of its control signal S7 in Figure 4. As in Figure 2, charge is transferred into the stage between ti and t2, causing V2 to fall in proportion to the incoming charge, settling at voltage 42. The change in V2 due to incoming charge is inversely proportional to the total capacitance of node 2 as explained above. In Figure 3 (neglecting parasitic capacitance) this total capacitance is C = C5 + C6.
After the charge is transferred in, the new features of Figure 3 come into play. At time t3A voltage VQRI conditionally switches from its high state 47 to low state 48. This conditional transition of VQRI is coupled via C6 to node 2 where, because of capacitive division, it produces a similar but smaller voltage change. The voltage at node 2 changes to voltage 49 (dashed line) if VQRI switches, and remains at voltage 42 (solid line) if it does not.
At time t3, Vci switches from high voltage 45 to low voltage 46, instigating charge transfer out of the stage. As explained with reference to Figure 2, node 2 is driven to a lower voltage due to coupling via capacitor 5. Charge-transfer circuit 3
removes charge from node 2 and transfers it to the next stage. By t4 V2 settles to voltage 43 which is independent of the charge previously on node 2, and charge transfer out of the stage is complete.
At t5 both Vci and VQRI return to their initial high states (voltages 45 and 47 respectively). This transition is identical for Va in every clock cycle. VQRI , however, may already be at its high voltage 47, depending on whether or not it switched at :3A. Thus the positive step coupled to node 2 at t5 can have different values, depending on the state Of VQR1, resulting in a different final voltage. The added switch 7 in Figure 3 is used to restore the voltage (and charge) on node 2 to a repeatable state regardless of the state of VQRI at t5. Switch 7 is turned on, as indicated by the high state of its control signal S7, during t5 - 16, thus establishing a ' repeatable voltage at node 2 to begin the next cycle, so voltage 44 = voltage 41. With an ideal switch, voltage 44 = Vp; practical MOS switches introduce a small 'pedestal' so that voltage 44 ≠ Vp. This non-ideality is, however, repeatable cycle- to-cycle, so the voltage 44 = voltage 41 condition is still met in practical circuits.
Unlike the case of Figure 1 where the charge transferred into the stage was subsequently transferred out without alteration, the outgoing charge in the circuit of Figure 3 differs in general from the incoming charge:
QOUT = QIN + C6ΔVQR1 + QCONST Equation 1
where C6 is the capacitance of capacitor 6, ΔVQRI is the change in VQR] at t3A, and QCONST is given by:
QCONST = (C5+C6)(voltage 41 - voltage 43) + C5(voltage 46 - voltage 45)
Equation IA
QCONST is nominally a fixed charge, since voltages 41, 43, 45, and 46 are all ideally constant. Departures from this ideal case, which constitute one source of charge-transfer imperfection, will be discussed below.
As is apparent in Figure 4, ΔVQRI is equal to (voltage 48 - voltage 47) if VQRI switches, and is equal to zero if it does not. Note that both charge quantities COΔVQRI and QCONST can be made either positive or negative by appropriate choices of the various voltages.
When the circuit of Figure 3 is used to form one stage of a pipelined ADC, the quantity (voltage 48 — voltage 47) is made equal to a reference voltage; for convenience it will be called VRJ . Correspondingly, the quantity C6VRI becomes a reference charge, since C6 is fixed in a given instantiation. Thus the conditional choice of ΔVQRI=VRI or ΔVQRI=0 at t3A corresponds in Equation 1 to the conditional addition of a reference charge C6VRI to the incoming charge packet Q1N. The circuit of Figure 3 thus provides one of the two operations required for a charge- domain ADC implementation.
Note that the exact position of time t3A is not critical to the operation of the circuit of Figure 3. The conditional transition of VQRI can occur at any time between t0 and t3 with no change in circuit performance; under some practicable conditions it may also occur in the t3 - U interval.
In some ADC implementations it is desirable to provide more than one conditional charge addition in a single pipeline stage. An example of such a stage is shown in Figure 5. This circuit includes, in addition to the elements in Figure 3, additional capacitor 6 A and voltage source VQR2. The operation of such a stage is identical to that of Figure 3, except that at t3A each of the voltages VQRI and VQR2 undergoes an independent conditional transition, of size VR1 and VR2 (or zero) respectively. The resulting charge transfer function of the stage is given by:
QOUT = QIN + C6 ΔVQRI + C6A ΔVQR2 + QCONST2 Equation 2
where QCONST2 is given by:
QCONST2 = (C5+C6+C6A)(voltage 41-voltage 43) + C5(voltage 46 - voltage 45)
Equation 2A
The same principle can be extended to any number of capacitors and VR values. For simplicity Equations 1 and IA will be used as the basis for the following discussion. The principles described are equally applicable to circuits with more than one conditionally-switched capacitor, as in Figure 5.
Two idealizations included in the discussion above are, in general, imperfectly realized in practical circuits: first, because of tolerances in manufacturing, conditionally-switched capacitors such as C6 generally do not have precisely the intended values; second, the final voltage to which the floating diffusion 2 settles (voltage 43 in Figure 4, for example) is in general not perfectly independent of QOUT- The effects of these non-idealities are considered in detail below, beginning with the dependence of voltage 43 on QOUT-
Considering a first-order (linear) dependence of voltage 43 upon QOUT, the value of voltage 43 can be written V43 = V43N + kQouτ, where V43 is the actual value of voltage 43, V23N is the nominal value, and k is a coefficient embodying the linear dependence on QOUT- Using this expression for voltage 43 in Equation IA yields:
QCONST = (C5+C6)(voltage 41 - v43N - kQOuτ) + C5(voltage 46 - voltage 45)
= (C5+C6)(voltage 41 - v43N) + C5(voltage 46 - voltage 45) - (C5+C6)kQOuτ
= Qcc - (C5+C6)kQOuτ Equation 3
where Qcc is the Qouτ-independent (i.e., constant) component of QCONST- Replacing QCONST in Equation 1 with the expression given by Equation 3, we obtain:
QOUT = QIN + C6AV0R1 + Qcc - (C5+C6)kQOuτ Equation 4
Turning now to the fabrication errors in C6, we can write C6 - C6N + C6E, where C6N is the nominal value of C6, and C6E is the deviation from nominal. Substituting this expression into Equation 4 yields:
QOUT = QIN + C6NΔVQRI + C6EΔVQRI + Qcc - (C5+C6)kQouτ Equation 5
In fractional terms, the errors expressed in Equation 5 are C6E/C6N and (C5+C6)k, both of which are dimensionless quantities. In practical designs these fractional errors are small (i.e., « 1). Thus we can find a practically-useful approximation to Equation 5 by replacing QOUT in the last term with the full expression given by Equation 5, and then omitting second-order error effects (i.e., terms including squares or products of the fractional errors). Defining ε = (C5+C6)k and carrying out this procedure, we obtain:
QOUT = QIN(1 -B) + C6NAVQR1 ( 1 -8) + QCC( 1 -S) + C6EAVQR1 Equation 6
Comparing this expression to the idealized expression of Equation 1 , it is apparent that the quantity (1-ε) is an effective charge-transfer gain; ε is the amount by which that gain falls short of unity. The term QEΔVQRI embodies the effect of fabrication error in C6
A pipelined charge-domain ADC is composed of multiple stages like that of Figure 3, in which reference charges are conditionally added and subsequently transferred down the pipeline. (As mentioned above, some architectures employ multiple conditional charges per stage.) Equation 6 expresses the output charge of each such stage in terms of that stage's input charge and its conditional capacitor switching. At any stage of the pipeline, the input charge is the sum of the signal- charge input to the pipeline and the cumulative changes due to up-stream pipeline stages according to Equation 6.
Consider for example the two-stage pipeline segment shown in Figure 6, which is composed of stages like that in Figure 3. The two stages, 61 and 62, each
consist of a storage node (67 and 68 respectively), a charge-transfer circuit (65 and 66 respectively), a conditionally-switched capacitor (63 and 64 respectively), a clocked capacitor (601 and 602), and a precharge switch Vp. The input charge to the pipeline segment, QPIN, is supplied to storage node 67 of stage 61 as indicated. The output of this pipeline segment (which is the output charge of stage 62) appears at node 69. The conditionally-switched voltages driving capacitors 63 and 64 are VQRI and VQR2 respectively. They can each independently be switched, with a step-size of 0 or VRI and 0 or VR2 respectively, where VR] and VR2 are reference voltages.
Such a pipeline operates in two-phases: alternating stages operate on alternating half-cycles of the clock. In the circuit of Figure 6, for example, charge is transfered into stage 61 and out of stage 62 on the first half-cycle, and is transferred out of stage 61 and into stage 62 on the second half-cycle. Details of this clocking method are not germane to the subject of the present invention, and are not treated further.
The input charge to stage 62 in Figure 6 is the output charge from stage 61. Thus, with the input charge to stage 61 equal to QPIN, the output charge from stage 62 can be derived by applying Equation 6 twice:
Qouτ62 = QiN62(l-ε) + C64NΔVQR2(l-ε) + Qcc(l-ε) + C64EΔVQR2
= [QP1N( I -S) + C63NΔVQRi(l-ε) + Qcc(l-ε) + C63EΔVQR1](l-ε)
+ C64NΔ VQR2( 1 -ε) + Qcc( 1 -ε) + C64EΔ VQR2
= QPINO -S)2 + C63NΔVQR1(l-ε)2 + C64NΔVQR2(l-ε)
+ C63EΔVQR!(l-ε) + C64EΔVQR2 + [(1-ε)2 + (l-ε)]QCC Equation 7
This expression can be simplified by omitting second-order error terms as was done above, giving:
Qouτ62 = QPIN(1 -2E) + C63NΔVQRi(l-2ε) + C64NΔVQR2(l-ε)
+ C63EΔVQRI + C64EΔVQR2 + (2-3ε)Qcc Equation 8
Equation 8 shows the cumulative effect of charge -transfer gain and capacitor errors in a two-stage pipeline. The same analysis can be extended to multiple stages and to multiple conditionally-switched capacitors per stage.
In order for a pipelined charge-domain ADC to produce linear results, it is essential that the conditionally-added charges from each stage appear at the end of the ADC pipeline with a specific ratio. In the Equation 8 the (non-zero) values of the conditionally-added charges are nominally C63NVRI and C64NVR2 in the first and second stages respectively. According to equation 8 then, the conditionally-added charge values, as they appear at the pipeline output, are:
[C63NVRI (1 -2S)+C63EVRI] from the first stage, and
[C64NVR2(I -ε)+C64 EVR2] from the second stage.
Thus the ADC linearity requirement can be expressed as:
[C63NVR1 (l-2ε)+C63EVR,] / [C64NVR2(I -8)+C64EVR2] = K Equation 9
where K is the intended ratio. In an ideal pipeline segment, in which the charge- transfer and capacitor errors are zero, Equation 9 is simplified to:
(C63NVR1) / (C64NVR2) = K Equation 10
The effects of non-zero gain error ε and capacitor errors such as C63E are evident when Equation 9 is compared to Equation 10.
One aspect of the present invention provides a way to satisfy Equation 9 when these errors have non-zero values. This can consist of providing an adjustment to the reference voltages VRI and VR2. The nominal capacitance values C63N and C64N and the reference voltages VRI and VR2 are chosen to satisfy Equation 5 10; then independently adjustable voltages VAi and VA2 are added to VRI and VR2.
With this change, and again omitting second-order error terms, the upper and lower terms in the ratio of Equation 9 become:
[C63NVR1(l-2ε) + C63EVR1] → [C63N(VR1+VA1)(l-2ε) + C63EVR1]
10
= [C63NVRi(l-ε) - SC63NVRI + C63EVR] + C63NVAI] and
[C64NVR2(I ^) + C64EVR2] → [C64N(VR2-HVA2)( I -E) + C64E VR2] I 5
= [C64NVR2 (1-ε) + C64EVR2 + C64NVA2)]
The added voltages are now adjusted to force the ratio of the adjusted bracketed terms to equal K. For example, setting: 0
VAI = [ε - (C63E/C63N)]VRi Equation 1 IA and
VA2 = - (C64E/C64N)VR2 Equation 1 1 B 5 results in the desired ratio. With this solution, VA2 is adjusted to correct for the error of capacitor 64 and VAI is adjusted to compensate both for the error of capacitor 63 and for the charge-transfer gain ε.
An alternative adjustment is: 0
VAI = (C63E/C63N)VRI Equation 12A and
VA2 = [- ε - (C64EZC64N)]VR2 Equation 12B
which also results in the desired ratio. With this solution, VAi is adjusted to correct for the error of capacitor 63 and VA2 is adjusted to compensate both for the error of capacitor 63 and for the charge-transfer gain ε. Both solutions 1 1 AJB and 12A/B are useful. Any linear combination of these solutions can be used with the same result.
This same adjustment principle can be applied in a pipeline with any number of stages. It can also be applied in an ADC design with more than one conditionally-added charge per stage (as in the example of Figure 5). In these cases, a separate adjustment voltage (VA) is applied to the reference voltage for each conditionally-switched capacitor. With the adjustment method of Equation 12A/B, for example, each such adjustment corrects for the combination of that individual capacitor's error and the total charge-transfer errors of all previous stages. In some designs, the charge-transfer error is not identical from stage to stage, but this adjustment method intrinsically accommodates such variation.
In addition, the same adjustment principle can be applied to a differential pipeline stage; in such an instance it may be preferable to generate a single adjustment voltage VA that is shared between the two members of the differential circuit.
Recall from the discussion of Figures 3 and 4 that the conditional voltage transition ΔVQRI which produces the conditional charge addition C6ΔVQRI results from switching VQRI between two fixed voltages (47 and 48 in Figure 4). This voltage difference constitutes the reference voltage VR. A practical way of adding the adjustment voltage VA to VR is to insert a small adjustable voltage in series with either voltage 47 or voltage 48. One realistic implementation of the driver and adjustable voltage source, 8 and 9 in Figure 11, is shown in Figure 7.
Figure 1 1 shows an example of a pipeline stage like that of Figure 1 with such an adjustment voltage added. Pipeline elements 1, 2, 3, 5, and 6 are equivalent
to those in Figure 1. The added elements are a driver 8 for conditionally-switched capacitor 6, and an adjustable voltage source 9. The output of driver 8 switches between supplied voltage V8 and the voltage at node 10 in response to a digital control signal SQRI , thus providing the required conditional voltage transition. Thus VH and voltage 10 provide the voltages identified as 47 and 48 respectively in
Figure 4. Adjustable voltage source 9 adjusts the low voltage supplied to capacitor 6 as described above.
Figure 12 shows a differential pipeline with similar function. Each member of the differential pair of pipeline nodes, 2 and 122, is provided with a conditionally- switched capacitor, 6 and 126 respectively. The voltage transitions driving capacitors 6 and 126 are provided by drivers 8 and 128, switching between VH and the voltages of nodes 10 and 130, in response to digital control signals SQRIA and SQRIB, respectively. The voltages of nodes 10 and 130 are supplied by adjustable voltage sources 9 and 129 respectively. Nodes 10 and 130 may be connected to form a single node, and supplied by a single adjustable source such as 9.
Figure 7 shows a CMOS inverter consisting of PFET 71 and NFET 72, supplied by fixed voltages VH and VL. The NFET is connected to VL via resistor 73 at node 75. An adjustable current 74 is also connected to node 75. The inverter is driven by a logic signal SQR I, and its output constitutes the voltage signal VQRI shown in Figure 3.
Figure 8 shows operating waveforms for the circuit of Figure 7. At time t0 logic signal SQRI is at a low state, turning PFET 71 on and NFET 72 off. Voltage
VQRI is at an initial high value which is equal to VH- At time t3A SQR I switches to a high logic state, turning FET 71 off and FET 72 on. FET 72 thus connects the output VQRI to node 75. If the adjustable current 74 is set to zero, then the series combination of FET 72 and resistor 73 charges VQRI towards VL, as indicated by voltage curve 81 in Figure 8. Since the load on this circuit consists only of a capacitor (capacitor 6 in Figure 3), there is no DC current through FET 72 or resistor 73, so VQR I eventually settles to VL. Thus the voltage transition of VQR I is VL-VH. This quantity constitutes the unadjusted reference voltage VR.
If the current source 74 is adjusted to a non-zero value IA, then the initial value of node 75 is VL+ VA = VL+IAR-73, where R73 is the value of resistor 73. (Note that FET 72 is initially off, so no current other than IA initially flows into resistor 73.) When SQRI changes state, then FET 72 turns on and connects the load capacitor to node 75, causing VQRI to charge downward along curve 82. At the end of this transition, current through FET 72 falls to zero and VQRI settles to final voltage VL+VA. Thus the voltage transition of VQRI is VL+VA-VH = VR+VA. Adjustable current source 74, which is easily realizable in a practical circuit, thus provides for adjusting the size of the transition in VQRI , as required. A similar circuit in which the resistor is placed in the source of PFET 71 instead of NFET 72 is equally practical. These circuits provide the necessary adjustment of VR with low power and small circuit area consumption.
As discussed in connection with Equations 1 IA and 12B, the required VA values have two components: one which corrects for a capacitor error and one which corrects for charge-transfer gain error. In the circuit of Figure 7, the current source 74 can be made to consist of two independent sources in parallel, each controlled independently to correct for one of the errors. The combined currents sum to develop the composite VA value.
The capacitor-error component corrected by VA can be expected to be temperature-invariant because it is due primarily to geometric variation between capacitors, which occurs during circuit fabrication but does not generally change thereafter. Thus an adjustment voltage VA which tracks VR provides a temperature- stable adjustment. Creating a component of VA which tracks VR over temperature using circuits similar to Figure 7 is straightforward in a conventional CMOS process. The value of this adjustment voltage can be set by a calibration process carried out during manufacturing test or upon powering-up the circuit, for example.
The second VA component corrects for charge-transfer gain error. This error depends on details of the charge -transfer circuits employed, and depends in general
on both fabrication-process variation and on operating temperature. Known BBD charge-transfer circuits include both conventional (passive) ones and ones employing active circuitry, such as those described in a previous patent application by the same inventor entitled "Boosted Charge-Transfer Pipeline", U.S. Patent Application No. 11/807,914, filed May 30, 2007, which is hereby incorporated by reference in its entirety. These charge-transfer circuits exhibit both dynamic and static components of charge-transfer gain error. One aspect of the present invention provides for generating an adjustment- voltage component which tracks the static component of charge-transfer error.
In the discussion leading to Equations 3-6, the charge-transfer gain error was formulated in terms of the variation of charge-transfer-circuit input voltage with the amount of charge transferred. In that discussion the final value of the voltage V43 at the input of the charge-transfer circuit was given as V43 = V43N + kQouτ, with V23N the nominal value, and k a coefficient embodying the.linear- dependence on Qoyτ- This formulation encompasses both dyanmic dependence of V43 on QOUT (due to incomplete settling) and static dependence. For the following discussion, it will be assumed that the dynamic effect is negligible, and the coefficient k reflects only static dependence.
It is known that the primary mechanism causing this static dependence is a voltage-feedback effect by which a voltage change at the charge-transfer circuit output causes its input voltage to change. As shown in Figure 2, the output voltage of a charge transfer circuit is a function of the output charge; thus the causation giving rise to the coefficient k is:
output charge → output voltage change → input voltage change
If we denote the coefficient relating output voltage change to input voltage change:
β = dvπVdvouT
- iy -
then the coefficient k relating output charge change to input voltage change is:
k = dvrN/dQouT = (dvrN/dvOuτ)(dvOuτ/dQouτ) = β dvOuτ/dQouτ
But d(vouτ)/d(Qouτ) is simply the inverse of the capacitance at the charge- transfer circuit output node (which is equal to the node capacitance of the next pipeline stage). Defining that capacitance as COUT, we then have:
k = β/Couτ
Referring again to Figures 3 and 4, we recall that the charge-transfer error ε was given, in terms of k and the storage-node capacitance C5+C6, as ε = (Cs+C6)k, leading to the expression:
ε = (C5+C6)k = (C5+C6)β/COuτ = β [(C5+C6)/COuτ] Equation 13
Thus the charge-transfer gain error ε depends on the voltage-feedback coefficient β of the charge-transfer circuit and a ratio of pipeline node capacitances. Since the pipeline node-capacitance ratios are known by design within small tolerances (discussed above), the gain error value ε can be derived from a determination of the voltage-feedback coefficient β.
Figure 9 shows a circuit which senses the voltage-feedback coefficient β. It consists of a charge-transfer circuit 91 with input node 92 and output node 93, a voltage source 95 connected to node 93, and a current source 94 connected to node 92. Charge-transfer circuit 91 is a replica of the charge-transfer circuits employed in the actual charge pipeline. Current source 94 is configured to sink from node 92 a small current which is typical of current levels near the end of the charge-transfer process described above. Voltage source 95 provides a voltage at node 93 which is in the range normally occurring at the output of the charge-transfer circuit at the end of the charge transfer process (such as voltage 42 in Figure 4). Thus the charge-
transfer circuit in Figure 9 is biased in a static condition essentially like its instantaneous condition near the end of the normal (clocked) charge-transfer process.
Figure 10 shows the voltages of the two nodes in Figure 9. V93, the voltage of node 93, is driven cyclically by voltage source 95 between two levels 101 and 102 differing by ΔV93. Due to the voltage-feedback effect discussed above, the input voltage of the charge-transfer circuit (i.e., the voltage V92 at node 92) responds by changing between two levels 103 and 104, with difference ΔV92. By the definition above, these voltage changes are related as:
ΔV92 = β ΔV93
Thus for a known (fixed) value of the drive voltage change ΔV93-, ΔV92 provides a direct measure of β. Using the known-by-design values of (C5+C6)/Couτ and the reference voltage, the appropriate adjustment voltage (VA2 in Equation 12B, for example) can be generated using known circuit techniques.
The voltage change ΔV92 can be converted to a DC voltage using phase- sensitive detection, since the alternating drive voltage V93 is available as a reference.
The frequency of this alternating voltage is not critical, and need not be as high as the sample rate of the charge pipeline, as the parameter being sensed changes only slowly (primarily due to chip temperature changes). Because of the low current at node 92, settling of V92 in response to each V93 transition is relatively slow, so the operating frequency of this circuit must be limited in order to obtain a valid settled value for ΔV92.
Alternatively, two circuits like that of Figure 9, with differing DC voltages supplied by the output-node voltage sources, thus producing two different input- node voltages, can be used to directly generate a static (DC) value of ΔV92. The alternating-voltage method described above is generally more accurate, however, since ΔV92 is quite small (typically only a few mV or tens of mV).
In a practical charge pipeline, the charge-transfer circuits may not be of identical design at all stages. The β-sensing circuitry just described consumes very little power, and can practically be reproduced for each charge-transfer circuit design employed.
The circuitry described provides a correction voltage based on a charge- transfer circuit which is a replica of the charge transfer circuits in the pipeline. Such a replica-based method provides very good tracking over operating conditions, but typically has small initial mismatches. Such initial mismatches can be removed in a calibration operation, at manufacturing test or during circuit power-up for example. After the calibration, the replica-based circuit provides tracking of subsequent changes in operating conditions, including temperature and supply voltage.
Such an initial calibration step also (simultaneously) provides correction for any dynamic component of charge-transfer gain error which is present under the calibration conditions. Any change of this dynamic error with operating-condition changes (especially temperature) after calibration, however, is not corrected by the techniques of this invention.
While this invention has been particularly shown and described with references to example embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the scope of the invention encompassed by the appended claims.
Claims
1. A pipelined charge domain circuit using bucket brigade charge transfer comprising: a first charge transfer circuit; a second charge transfer circuit; a node coupled to the first charge transfer circuit and the second charge transfer circuit; a clocked capacitor coupled to the node and to a clocked voltage; a conditionally-switched capacitor coupled to the node, the conditionally- switched capacitor driven by a transition voltage; and an adjustment circuit, for adjusting the transition voltage according to conditions within the pipelined charge domain circuit.
2. The pipelined charge-domain circuit of claim 1 wherein the condition is an incorrect conditionally-added charge.
3. The pipelined charge-domain circuit of claim 1 wherein the condition is non- optimum charge transfer gain.
4. The pipelined charge-domain circuit of claim 3 wherein the condition is sub- unity charge transfer gain.
5. The pipelined charge-domain circuit of claim 1 wherein the condition comprises both an incorrect conditionally-added charge and non-optimum charge transfer gain. - 15 -
6. The pipelined charge-domain circuit of claim 1 wherein at least one of the first or second charge transfer circuit is a boosted charge transfer circuit.
7. The pipelined charge-domain circuit of claim 1 , further comprising: two or more charge transfer stages arranged in a pipeline,; and wherein the adjustment circuit for a given converter stage in the pipeline adjusts the transition voltage according to non-ideal charge transfer gain in at least one preceding pipeline stage.
8. The pipelined charge-domain circuit of claim 7 wherein the adjustment circuit for a given converter stage adjusts the transition voltage according to non- unity charge transfer gain in two or more preceding pipeline stages.
9. The pipelined charge-domain circuit of claim 1 wherein the adjustment circuit further comprises: an adjustable current source having an output node; a resistor, coupled between a reference voltage and the output node of the current source; and a switching Field Effect Transistor (FET) coupled between the output node of the adjustable current source and the transition voltage.
10. The charge-domain analog-to-digital converter of claim 7 wherein at least two converter stages each comprise an adjustment circuit; and wherein a first adjustment voltage component of the transition voltage corrects for both an error introduced by a respective conditionally switched capacitor and an error introduced by charge transfer associated with its respective stage; a second adjustment voltage component of the transition voltage corrects for at least an error introduced by a respective conditionally switched capacitor associated with its respective stage.
1 1. The pipelined charge domain circuit of claim 1 operating as an analog-to- digital converter and additionally comprising: a precharge switch, coupled to the node, for providing a precharge voltage.
12. The pipelined charge domain circuit of claim 1 wherein the condition is an estimate of a voltage-feedback coefficient of the first charge transfer circuit.
13. The pipelined charge domain circuit of claim 12 wherein the condition is inferred from a replica circuit for at least one of the first or second charge transfer circuits.
14. The pipelined charge domain circuit of claim 13 wherein the adjustment circuit further couples an alternating voltage to an output terminal of a replica of the first charge transfer circuit and further comprises: a current source, for coupling a predetermined direct current to an input terminal of the replica circuit; and a detector, for synchronously detecting a resulting voltage change at the input terminal of the replica circuit.
15. The pipelined charge domain circuit of claim 13 wherein the adjustment circuit further couples two different direct current voltages to output terminals of two replica circuits for a selected charge transfer circuit and comprises: current sources, for causing a predetermined current flow to input terminals of the replica circuits; and a sensor, for sensing a voltage difference at the input terminals of the replica circuits.
16. The charge domain pipeline of claim 1 arranged to provide a differential charge domain pipeline and further comprising: a third charge transfer circuit; a fourth charge transfer circuit; a second node coupled to the third charge transfer circuit and the fourth charge transfer circuit; a second clocked capacitor coupled to the second node and to a second clocked voltage; and a second conditionally-switched capacitor coupled to the second node, the second conditionally-switched capacitor driven by an adjustable transition voltage; and
a third and fourth capacitor coupled to the first node and second node respectively and configured to provide conditional charge to either the first charge transfer circuit or the third charge transfer circuit.
17. The charge domain pipeline of claim 16 wherein the adjustment circuit couples the same transition voltage to both the first and second conditionally switched capacitors.
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020097015449A KR101460818B1 (en) | 2007-01-23 | 2008-01-23 | Analog error correction for a pipelined charge-domain a/d converter |
EP08724712.8A EP2106586B1 (en) | 2007-01-23 | 2008-01-23 | Analog error correction for a pipelined charge-domain a/d converter |
CN2008800026409A CN101622602B (en) | 2007-01-23 | 2008-01-23 | Analog error correction for a pipelined charge-domain a/d converter |
KR1020080028761A KR20090103266A (en) | 2007-01-23 | 2008-03-28 | Enhanced Functional Herb Medicine Dentifrice and Its Production Methods |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US88196707P | 2007-01-23 | 2007-01-23 | |
US60/881,967 | 2007-01-23 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2008091603A1 true WO2008091603A1 (en) | 2008-07-31 |
Family
ID=39640700
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/US2008/000825 WO2008091603A1 (en) | 2007-01-23 | 2008-01-23 | Analog error correction for a pipelined charge-domain a/d converter |
Country Status (5)
Country | Link |
---|---|
US (2) | US7535400B2 (en) |
EP (1) | EP2106586B1 (en) |
KR (2) | KR101460818B1 (en) |
CN (1) | CN101622602B (en) |
WO (1) | WO2008091603A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8344927B2 (en) | 2010-09-15 | 2013-01-01 | Electronics And Telecommunications Research Institute | Analog digital converting device and reference voltage controlling method thereof |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8766839B2 (en) * | 2012-09-07 | 2014-07-01 | Texas Instruments Incorporated | Reducing the effect of elements mismatch in a SAR ADC |
KR20180015062A (en) * | 2016-08-02 | 2018-02-12 | 주식회사 피글 | Composition for periodontal disease and toothpaste comprising the same |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5149954A (en) * | 1991-03-26 | 1992-09-22 | Santa Barbara Research Center | Hold capacitor time delay and integration with equilibrating means |
US20040160351A1 (en) * | 2002-08-23 | 2004-08-19 | Giuseppe Rossi | Fully differential reference driver for pipeline analog to digital converter |
US6867718B2 (en) * | 2003-01-24 | 2005-03-15 | Stmicroelectronics S.R.L. | Method of correction of the error introduced by a multibit DAC incorporated in an ADC |
US20050151679A1 (en) * | 2003-12-12 | 2005-07-14 | Acqiris | ADC with digital error correction |
Family Cites Families (39)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5339275A (en) | 1970-12-28 | 1994-08-16 | Hyatt Gilbert P | Analog memory system |
US3801826A (en) * | 1972-05-12 | 1974-04-02 | Teletype Corp | Input for shift registers |
DE2317253C3 (en) * | 1973-04-06 | 1975-09-25 | Deutsche Itt Industries Gmbh, 7800 Freiburg | Bucket chain control |
US3969636A (en) | 1975-06-30 | 1976-07-13 | General Electric Company | Charge sensing circuit for charge transfer devices |
US4072938A (en) | 1975-08-25 | 1978-02-07 | Westinghouse Electric Corporation | Bucket brigade analog-to-digital converter |
NL7610351A (en) | 1976-09-17 | 1978-03-21 | Philips Nv | LOAD TRANSFER DEVICE. |
US4138665A (en) | 1977-09-21 | 1979-02-06 | General Electric Company | Preamplifier for analog to digital converters |
US4224585A (en) | 1978-08-28 | 1980-09-23 | Reticon Corporation | Methods and apparatus for compensating for charge transfer inefficiency in imaging and other variable length charge transfer devices |
US4538287A (en) | 1979-06-04 | 1985-08-27 | Texas Instruments Incorporated | Floating gate amplifier using conductive coupling for charge coupled devices |
NL8102100A (en) * | 1981-04-29 | 1982-11-16 | Philips Nv | COMPENSATION OF 1ST ORDER EFFECT OF TRANSPORT LOSS IN A C.T.D. |
US4471341A (en) * | 1982-03-03 | 1984-09-11 | Rca Corporation | Pipe-lined CCD analog-to-digital converter |
JPS59963A (en) | 1982-06-28 | 1984-01-06 | Hitachi Ltd | Charge transfer device |
JPS5921059A (en) * | 1982-07-28 | 1984-02-02 | Hitachi Ltd | Charge transfer circuit |
KR930007280B1 (en) | 1983-09-07 | 1993-08-04 | 가부시기가이샤 히다찌세이사꾸쇼 | Semiconductor memory having charge transfer device voltage-amplifier |
FR2566538B1 (en) | 1984-06-26 | 1986-11-14 | Thomson Csf | ANALOG COMPARATOR WITH LOAD TRANSFER AND DEVICES USING SUCH A COMPARATOR |
JPH079981B2 (en) | 1985-02-05 | 1995-02-01 | ソニー株式会社 | Charge transfer device |
US4984256A (en) * | 1987-02-13 | 1991-01-08 | Kabushiki Kaisha Toshiba | Charge transfer device with booster circuit |
JPH0728029B2 (en) * | 1987-05-21 | 1995-03-29 | 株式会社東芝 | Charge transfer device |
NL8701528A (en) | 1987-06-30 | 1989-01-16 | Philips Nv | SEMICONDUCTOR DEVICE FITTED WITH A LOAD TRANSFER. |
JP2768736B2 (en) | 1989-05-29 | 1998-06-25 | 株式会社東芝 | Charge transfer device |
FR2657719B1 (en) | 1990-01-30 | 1994-08-26 | Thomson Composants Militaires | SAMPLING CIRCUIT FOR ANALOG SIGNALS. |
JP2666522B2 (en) | 1990-05-14 | 1997-10-22 | 日本電気株式会社 | Charge transfer device |
US5291083A (en) | 1993-01-12 | 1994-03-01 | Hewlett-Packard Company | Bucket brigade analog delay line with voltage limiting feedback |
US5349303A (en) | 1993-07-02 | 1994-09-20 | Cirque Corporation | Electrical charge transfer apparatus |
US5671252A (en) * | 1994-09-21 | 1997-09-23 | Analog Devices, Inc. | Sampled data read channel utilizing charge-coupled devices |
JPH09129864A (en) | 1995-10-30 | 1997-05-16 | Canon Inc | Semiconductor device, and semiconductor circuit, correlation operating device and signal processing system which use the semiconductor device |
US5880777A (en) | 1996-04-15 | 1999-03-09 | Massachusetts Institute Of Technology | Low-light-level imaging and image processing |
US5693968A (en) | 1996-07-10 | 1997-12-02 | Board Of Supervisors Of Louisiana State University And Agricultural And Mechanical College | Bi-directional, fast-timing, charge coupled device |
JP4066211B2 (en) | 1997-06-06 | 2008-03-26 | 財団法人国際科学振興財団 | Charge transfer amplifier circuit, voltage comparator and sense amplifier |
JP3164032B2 (en) * | 1997-10-01 | 2001-05-08 | 日本電気株式会社 | Boost circuit |
JP3529022B2 (en) | 1998-01-30 | 2004-05-24 | シャープ株式会社 | Charge transfer element |
US6291083B1 (en) * | 1999-01-25 | 2001-09-18 | Sanoh Kogyo Kabushiki Kaisha | Steel product with plating layers |
US6249181B1 (en) | 1999-11-30 | 2001-06-19 | William J. Marble | Differential-mode charge transfer amplifier |
US6356148B1 (en) | 1999-11-30 | 2002-03-12 | Ami Semiconductor, Inc. | Systems and methods for enhancing charge transfer amplifier gain |
US6617992B2 (en) * | 2001-08-15 | 2003-09-09 | National Semiconductor Corporation | Capacitor mismatch independent gain stage for differential pipeline analog to digital converters |
US6566943B1 (en) | 2001-12-26 | 2003-05-20 | Ami Semiconductor, Inc. | Reference-free charge transfer amplifier |
US6859159B2 (en) * | 2002-06-18 | 2005-02-22 | Analog Devices, Inc. | Switched-capacitor structures with enhanced isolation |
US6751141B1 (en) | 2002-11-26 | 2004-06-15 | Intel Corporation | Differential charge transfer sense amplifier |
US6972707B1 (en) * | 2004-07-12 | 2005-12-06 | Massachusetts Institute Of Technology | Sub-ranging pipelined charge-domain analog-to-digital converter with improved resolution and reduced power consumption |
-
2008
- 2008-01-23 US US12/011,029 patent/US7535400B2/en active Active
- 2008-01-23 WO PCT/US2008/000825 patent/WO2008091603A1/en active Search and Examination
- 2008-01-23 CN CN2008800026409A patent/CN101622602B/en not_active Expired - Fee Related
- 2008-01-23 KR KR1020097015449A patent/KR101460818B1/en not_active IP Right Cessation
- 2008-01-23 EP EP08724712.8A patent/EP2106586B1/en not_active Not-in-force
- 2008-03-28 KR KR1020080028761A patent/KR20090103266A/en not_active Application Discontinuation
-
2009
- 2009-04-14 US US12/423,652 patent/US7719456B2/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5149954A (en) * | 1991-03-26 | 1992-09-22 | Santa Barbara Research Center | Hold capacitor time delay and integration with equilibrating means |
US20040160351A1 (en) * | 2002-08-23 | 2004-08-19 | Giuseppe Rossi | Fully differential reference driver for pipeline analog to digital converter |
US6867718B2 (en) * | 2003-01-24 | 2005-03-15 | Stmicroelectronics S.R.L. | Method of correction of the error introduced by a multibit DAC incorporated in an ADC |
US20050151679A1 (en) * | 2003-12-12 | 2005-07-14 | Acqiris | ADC with digital error correction |
Non-Patent Citations (1)
Title |
---|
See also references of EP2106586A4 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8344927B2 (en) | 2010-09-15 | 2013-01-01 | Electronics And Telecommunications Research Institute | Analog digital converting device and reference voltage controlling method thereof |
Also Published As
Publication number | Publication date |
---|---|
EP2106586A4 (en) | 2012-02-29 |
KR101460818B1 (en) | 2014-11-11 |
KR20090113266A (en) | 2009-10-29 |
CN101622602B (en) | 2012-01-04 |
EP2106586A1 (en) | 2009-10-07 |
US7719456B2 (en) | 2010-05-18 |
CN101622602A (en) | 2010-01-06 |
US20090201184A1 (en) | 2009-08-13 |
KR20090103266A (en) | 2009-10-01 |
US20080174466A1 (en) | 2008-07-24 |
EP2106586B1 (en) | 2014-11-12 |
US7535400B2 (en) | 2009-05-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Ahmed et al. | A low-power capacitive charge pump based pipelined ADC | |
US10826523B2 (en) | Analog-to-digital converter, measurement arrangement and method for analog-to-digital conversion | |
US20090201051A1 (en) | Sample-and-Hold Circuit and Pipeline Ad Converter Using Same | |
US20070103203A1 (en) | Sample hold circuit and multiplying D/A converter having the same | |
US20110205099A1 (en) | Successive approximation type a/d converter circuit | |
US10181857B1 (en) | Analog-to-digital converter error correction | |
US5872469A (en) | Switched capacitor circuit adapted to store charge on a sampling capacitor related to a sample for an analog signal voltage and to subsequently transfer such stored charge | |
CN112219097B (en) | Temperature sensor semiconductor device with paired diodes and feedback loop | |
USRE36014E (en) | Voltage storage circuits | |
JPS59132231A (en) | Analog-digital converter | |
US7830159B1 (en) | Capacitor mismatch measurement method for switched capacitor circuits | |
EP2127085B1 (en) | Charge-domain pipelined analog-to-digital converter | |
US7719456B2 (en) | Analog error correction for a pipelined charge-domain A/D converter | |
JPH06177765A (en) | Dielectric-relaxation correction circuit of electric-charge redistribution a/d converter | |
US7446573B1 (en) | Comparator systems and methods | |
CN111628773B (en) | Analog-to-digital converter and analog-to-digital conversion method | |
JPS61193521A (en) | Analog-digital converting circuit | |
US7576586B2 (en) | Common-mode charge control in a pipelined charge-domain signal-processing circuit | |
US20090212830A1 (en) | Buffer device for switched capacitance circuit | |
WO2024082732A1 (en) | Analog-to-digital converter integrated with reference voltage generation, and calibration method | |
RU2380829C2 (en) | Analogue-to-digital converter | |
US10026497B2 (en) | Charge injection noise reduction in sample-and-hold circuit | |
JP2002133891A (en) | Sample-and-hold circuit | |
JPH08107356A (en) | A/d converter | |
Miyazaki et al. | Low-Power Area-Efficient Pipelined A/D Converter Design Using a Single-Ended Amplifier |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 200880002640.9 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 08724712 Country of ref document: EP Kind code of ref document: A1 |
|
DPE1 | Request for preliminary examination filed after expiration of 19th month from priority date (pct application filed from 20040101) | ||
WWE | Wipo information: entry into national phase |
Ref document number: 1020097015449 Country of ref document: KR |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2008724712 Country of ref document: EP |
|
DPE1 | Request for preliminary examination filed after expiration of 19th month from priority date (pct application filed from 20040101) |