US 6,384,758 B1Grant
High-speed sampler structures and methods
Issue Date:2002-05-07
•20 Claims
•7 Drawing Sheets
Abstract
High-speed sampler methods and structures are provided to enhance the correlation between an input signal Sinand a corresponding sampler output voltage Vout. An input buffer is enabled during sampling time periods and disabled during holding time periods. In the sampling time periods, a sampling capacitor Csis directly charged through the input buffer and the capacitor's bottom plate to a charge that corresponds to the input signal Sin. In the holding time periods, the disabled input buffer is isolated from the sampling capacitor Csand a common-mode signal Scmis directly coupled to the capacitor's bottom plate to provide the output voltage Voutat the capacitor's top plate. Preferably, an output capacitor Cois coupled to the sampling capacitor Csand charge from the sampling capacitor Csis transferred to the output capacitor Co.
Metadata
Assignee
- Analog Devices, Inc.
Inventors
- Christopher Michalski
- David Graham Nairn
Application Information
Application Number:US 09/723,956
Filing Date:2000-11-27
Priority Date:2000-11-27
Art Unit:7
Classifications
IPC:
H03M 100
Field of Search:
341122341156341155341161341162327 91327 94327 95327 96307353
Patent Drawings (7 sheets)
Description
Background of the Invention
[0002] 1. Field of the Invention
[0003] The present invention relates generally to samplers that sample an input signal Sinand hold a corresponding output voltage Voutthrough successive and alternate sampling and holding time periods.
[0004] 2. Description of the Related Art
[0005] Signal samplers are typically used to successively sample and hold the amplitudes of analog signals and provide the resulting samples for further processing (e.g., in an analog-to-digital conversion). Sampler structures have been described under various names (e.g., sample and holds, sample-and-hold amplifiers (SHAs) and track and hold circuits) and various structural configurations have been proposed (e.g., see U.S. Pat. Nos. 4,962,325 issued Oct. 9, 1990 to Miller, et al., 5,315,170 issued May 25, 1994 to Vinn, et al., 5,389,929 issued Feb. 14, 1995 to Nayebi, et al., 5,418,408 issued May 23, 1995 to Mangelsdorf, et al., 5,457,418 issued Oct. 10, 1995 to Chang, 5,838,175 issued Nov. 17, 1998 to Hsieh and 6,028,459 issued Feb. 22, 2000 to Birdsall, et al.).
[0006] Although various conventional sampler configurations can sample and hold an input signal Sin, their structures often corrupt this process so that the correlation between the input signal Sinand a corresponding output voltage Voutis degraded. FIG. 3 of U.S. Pat. No. 4,962,325, for example, shows a switched-capacitor prior art sampler in which serially-arranged input switches gate an input signal to sampling capacitors that are coupled to the input of a differential output amplifier. Other switches facilitate transfer of charges from the sampling capacitors to output capacitors that are arranged across the differential output amplifier.
[0007] When the serially-arranged input switches are realized with high speed transistors, the input signal Sinis typically passed through the transistors' current terminals and this passage is gated with control signals on the transistors' control terminals. In this arrangement, however, the input switches corrupt the input signal Sinwith consequent degradation (e.g., harmonic distortion) of the sampler's output voltage Vout.
Summary of the Invention
[0008] The present invention is directed to sampler methods and structures that enhance the correlation between an input signal Sinand a corresponding sampler output voltage Vout. These goals are realized by providing a supply current to an input buffer to enable it during sampling time periods and removing the supply current to disable and isolate the input buffer during holding time periods.
[0009] In the sampling time periods, the bottom plate of a sampling capacitor Csis directly charged through the input buffer to a charge that corresponds to the input signal Sin.
[0010] In the holding time periods, the input buffer is isolated from the sampling capacitor Csand a common-mode signal Scmis directly coupled the capacitor's bottom plate to transfer electrical charge from the sampling capacitor Csto the output capacitor Coand thereby generate the output voltage Voutacross the output capacitor Co.
[0011] These processes facilitate the replacement of serially-arranged switches and their attendant harmonic distortion with an alternately enabled and disabled input buffer that alternately charges a sampling capacitor Csand is isolated from the sampling capacitor Cs. Sampler embodiments are described for practicing the processes of the invention.
[0012] The novel features of the invention are set forth with particularity in the appended claims. The invention will be best understood from the following description when read in conjunction with the accompanying drawings.
Brief Description of the Drawings
[0013] FIG. 1 is a flow chart that illustrates a sampling method embodiment of the present invention;
[0014] FIG. 2 is a schematic of a sampler embodiment for practicing the method of FIG. 1;
[0015] FIG. 3A is a schematic of a complementary metal-oxide semiconductor (CMOS) embodiment of the sampler of FIG. 2;
[0016] FIG. 3B is a schematic of a bipolar CMOS (BiCMOS) embodiment of the sampler of FIG. 2;
[0017] FIG. 4A is a schematic of a differential CMOS embodiment of the sampler of FIG. 2;
[0018] FIG. 4B is a schematic of a differential BiCMOS embodiment of the sampler of FIG. 2; and
[0019] FIG. 5 is a block diagram of a subranging analog-to-digital converter (ADC) that includes a sampler of the present invention.
Detailed Description of the Preferred Embodiments
[0020] FIG. 1 is a flow chart 20 that is directed to the sampling and holding of an input signal Sinthat includes a common-mode signal Scmcomponent (i.e., the input signal Sinis superimposed on the common-mode signal Scm). In particular, the flow chart recites a method of sampling the input signal Sinand holding a corresponding output voltage Voutthrough successive and alternate sampling and holding time periods. A first process step 22 of FIG. 1 provides a sampling capacitor Csand an output capacitor Coand successively alternates between sampling and holding time periods.
[0021] Decision 23 asks if the method is in a sampling time period and moves, if the answer is affirmative, to process steps 24 and 26. Process step 24 provides a supply current to enable an input buffer. In process step 26, the input signal Sincharges the sampling capacitor Csthrough the enabled input buffer to a charge that corresponds to the input signal Sin.
[0022] If the answer to decision 23 is negative, the method proceeds to process steps 30 and 32. Step 30 disables the input buffer by removing the supply current from it. In the holding time periods, step 32 couples the sampling capacitor Csbetween the common-mode signal Scmand the output capacitor Coto directly transfer electrical charge from the sampling capacitor Co. to the output capacitor Coand thereby generate the output voltage Voutacross the output capacitor Co.
[0023] The processes of FIG. 1 can be best understood in the context of sampler structures with which the processes may be practiced. For example, FIG. 2 is a schematic of a sampler 40 that includes an input buffer 42, a current-source circuit 44, a differential amplifier 46, an output capacitor 48 and a sampling capacitor 50.
[0024] The input buffer 42 responds to an input signal Sinat an input port 52 and the differential amplifier 46 provides a corresponding output voltage Voutat an output port 54. The sampling capacitor Cs(50) is directly connected to the output port 56 of the input buffer 42 and to the inverting input port 58 of the differential amplifier 46. The output capacitor Cois coupled between the inverting input port 58 and the output port 54. In addition, a charging switch 60 is positioned between the inverting input port 58 and the noninverting input port of the differential amplifier 46 (via signal grounds 59) and a charge-transfer switch 62 couples a bottom plate 63 of the sampling capacitor Csto the common-mode signal Scm.
[0025] During the sampling time periods of process step 22 of FIG. 1, the sampler 40 is placed in a sampling operational mode wherein a charging signal Schrgis applied to close the charging switch 60 and the charge-transfer switch (62) remains open. The current-source circuit 44 is arranged to provide a supply current 64 to the input buffer 42 so that it is enabled during the sampling time periods.
[0026] Accordingly, the enabled input buffer 42 responds to the input signal Sinby generating a charging current 65 that charges the sampling capacitor Cs(50) to a sampled electrical charge Qs=CsSinthat corresponds to the present value of the input signal Sin. Thus, the sampler 40 realizes the process steps 24 and 26 of FIG. 1 during the sampling time periods. Closing the charging switch 60 during the sampling time periods essentially connects the input signal Sin, the input buffer 42, the sampling capacitor Csand the charging switch 60 in a charging path that is defined by the charging current 65 in FIG. 1.
[0027] During the holding time periods of process step 22 of FIG. 1, the sampler 40 is placed in a holding operational mode wherein a transfer signal Strnsis applied to close the charge-transfer switch 62 (opposite to the position shown) and the charging signal Schrgis removed to thereby open the charging switch 60 (opposite to the position shown). The current-source circuit 44 is arranged to remove the supply current 63 from the input buffer 42 so that it is disabled. The input buffer is thus isolated from the succeeding elements of the sampler (in particular, from the sampling capacitor Cs) during the holding time periods.
[0028] Accordingly, the low-impedance common-mode signal Scmdirectly drives the bottom plate 63 of the sampling capacitor Csand a charge-transfer current 66 flows from the sampling capacitor Csto the output capacitor Co. The sampled electrical charge Qsis thus directly transferred from the sampling capacitor 50 to become an output electrical charge Qoin the output capacitor 48 wherein Qo=Qsbecause of charge conservation.
[0029] The output electrical charge generates the output voltage Voutacross the sampling capacitor Cs(50) so that
[0030] Qo=CoVout=CsSin (1)
[0031] and therefore
[0032] Vout=(Cs/Co)Sin (2)
[0033] That is, charge is transferred to generate an output voltage Voutacross the output capacitor Cowith a gain of Cs/Corelative to the input signal Sin.
[0034] The current-source circuit 44 of FIG. 2 includes a current source 70 that generates the supply current 64 and also includes an enable switch 72 that couples the current source to the input buffer 42. During the sampling time periods of the process step 22 of FIG. 1, an enable signal Senblcloses the enable switch 72 so that the supply current 64 is provided to enable the input buffer 42.
[0035] During the holding time periods of the process step 22 of FIG. 1, the enable signal Senblis removed to thereby open the enable switch 72 (opposite to the position shown) so that the supply current 64 is removed from the input buffer 42 and it is thereby disabled and isolated. Preferably, the current-source circuit 44 also includes a steering switch 74 that is closed (opposite to the position shown) by a steering signal Sstrso that the supply current 64 is steered to a supply voltage VDDduring the holding time periods.
[0036] The sampler 40 of FIG. 2 has a “bottom plate” architecture wherein the input buffer 42 is enabled and supplies charge directly to the bottom plate 63 of the sampling capacitor 50 in the sampling time periods and the common-mode signal Scmis directly connected to the bottom plate 63 during the holding time periods. In the latter periods, the buffer is isolated from the bottom plate 63 because it has been disabled.
[0037] The sampler 40 differs from a first group of conventional samplers that charge a sampling capacitor through its top plate (e.g., the top plate 76 of the sampling capacitor 50 in FIG. 2) and transfer charge also from this top plate. The sampler 40 also differs from a second group of conventional samplers that have a “bottom plate” architecture but insert a charging switch between the input signal Sinand the bottom plate.
[0038] This serially-arranged charging switch is typically closed during sampling time periods and opened during holding time periods. In an exemplary charging switch arrangement, the source S of a metal-oxide semiconductor (MOS) transistor is connected to the input signal Sin, the transistor's drain D is connected to the sampling capacitor's bottom plate and the transistor's gate G is connected to a switching signal.
[0039] This structure, unfortunately, causes the transistor's gate-to-source voltage VGSto vary with the input signal Sin. Because the transistor's parasitic capacitances (e.g., CGS, CGDand bulk capacitances CBDand CBS) vary non-linearly with voltage, significant harmonic distortion is thus generated in the output voltage Voutat relatively low input frequencies.
[0040] A third group of conventional samplers seek to reduce the transistor's distortion by adding supplemental structures that add a version of the input signal Sinto the switching signal at the transistor's gate. This additional structure tends to reduce the varying signal across the CGSand CGDparasitic capacitances but has no comparable “bootstrapping” effect for the bulk parasitic capacitances.
[0041] This third group of conventional samplers thus realizes some distortion reduction but suffers the complexity and degraded reliability that result from the additional sampler structures. Moreover, the performance of this sampler arrangement at high sampling rates (e.g., >100 megasamples/second) and high analog input frequencies has typically been unsatisfactory.
[0042] In contrast, the sampler 40 of FIG. 2 eliminates serially-arranged switching structures and, instead, directly drives the bottom plate 63 of the sampling capacitor 50 with the enabled input buffer 42 during sampling time periods and directly drives the bottom plate with the common-mode signal Scmduring holding time periods when the input buffer is disabled and isolated. The sampler 40 thereby avoids the distortion of serially-arranged switching structures and the complexity of additional input signal structures.
[0043] The output electrical charge Qomust be removed from the output capacitor 48 during each sampling time period so that it does not add to the sampled electrical charge Qsthat will be transferred to it in the succeeding holding time period. This removal insures that the output voltage Voutof equations (1) and (2) corresponds to the input signal Sinof the preceding sampling time period.
[0044] This discharge of the output capacitor Cocan be realized, for example, by coupling the output of the differential amplifier 46 to the ground 59 through a discharge switch Sd(77) as shown in FIG. 2. In operation, the discharge switch 77 is opened (e.g., by the charging signal Schrg) during the holding time periods. An equivalent discharge structure is assumed but not shown in other sampler embodiments of the invention. Alternatively, the discharge switch Sdcan be omitted and the structure of FIG. 2 used as an integrator (e.g., in a Delta-Sigma A/D converter).
[0045] Process step 32 of FIG. 1 coupled the sampling capacitor Csbetween the common-mode signal Scmand the output capacitor Coto directly transfer electrical charge from the sampling capacitor Csto the output capacitor Coand thereby generate an output voltage Voutacross the output capacitor Co. In other sampler embodiments, the sampling capacitor Csmay simply be coupled to the sampler output port 54 to provide a sampled output voltage Vout.
[0046] Preferably, a buffer is inserted to provide buffering between the sampling capacitor Csand circuits that are attached to the output port 54. This buffer may be realized, for example, by replacing the output capacitor Coin FIG. 2 with a switch 78 (shown in broken lines) which closes during the holding periods to convert the differential amplifier 46 into a voltage follower.
[0047] FIG. 3A illustrates a CMOS embodiment 80 of the sampler 40 of FIG. 2 in which like elements are indicated by like reference numbers. The sampler 80, however, realizes the input buffer 42 with a common source CMOS buffer transistor 82 and also realizes the charging switch 60, the charge-transfer switch 62, the enable switch 72 and the steering switch 74 with CMOS transistors.
[0048] In particular, the sampler 80 includes charging transistor 84, a charge-transfer transistor 86, an enable transistor 88 and a steering transistor 90. The enable transistor 88 and the steering transistor 90 form a differential pair of transistors within a current-source circuit 92 that steer the supply current 64 to the buffer transistor 82 during sampling time periods and to the supply voltage VDDduring holding time periods.
[0049] In the holding time periods, charge-transfer transistor 86 is on and brings the source of the input buffer transistor 82 near the value of the common-mode signal Scm. Accordingly, the common-mode signal Scmis selected sufficiently large enough to ensure that the input buffer transistor 82 never turns on (i.e., is enabled) for any value of the input signal Sinat its gate. The common-mode signal Scmmay be generated in various conventional ways. For example, a first resistive divider can be used to apply a selected common-mode signal Scmto the input signal Sinand a similar second resistive divider can provide an identical common-mode signal Scmto the charge-transfer transistor 86.
[0050] FIG. 3B illustrates a bipolar CMOS (BiCMOS) embodiment 100 of the sampler 40 of FIG. 2 in which like elements are indicated by like reference numbers. In contrast to the sampler 80 of FIG. 3A, the sampler 100 realizes the input buffer 42 of FIG. 2 with a common emitter bipolar buffer transistor 102, realizes the charging switch 60 with a CMOS charging transistor 104 and realizes the charge-transfer switch 62 with a bipolar charge-transfer transistor 106.
[0051] In the sampling time periods of the sampler 100, the charging signal Schrgis applied to bias on the charging transistor 104. During this time, the charge-transfer transistor 106 is off so that the input buffer transistor 102 is enabled by receipt of the supply current 64 from the current source 70.
[0052] In the holding time periods, the charge-transfer transistor 106 is turned on by the transfer signal Strnsso that the supply current 64 is steered through the charge-transfer transistor. Because the supply current is steered away from the input buffer transistor 102, this transistor is disabled and isolated. The transfer signal Strnsis applied to the base of the charge-transfer transistor 106 and its value is selected so that a signal substantially equal to the common-mode signal Scmappears at the emitter where it drives the bottom plate 63 of the sampling capacitor 50. Because the charge-transfer transistor 106 transfers charge from the sampling capacitor 50 to the output capacitor 48 and also disables the input buffer transistor 102, the signal on its base is titled both as a transfer signal Strnsand as a disable signal Sdsblin FIG. 3B.
[0053] It is apparent that the input buffer transistor 102 and the charge-transfer transistor 106 form a differential pair of transistors that are each disabled when the supply current 64 is steered to the other of the differential pair. The common-mode signal Scmon the base of the charge-transfer transistor 106 is selected sufficiently large enough to ensure that the input buffer transistor 102 never turns on (i.e., be enabled) for any value of the input signal Sinat its base.
[0054] FIG. 4A illustrates a differential CMOS embodiment 120 of the sampler 80 of FIG. 3A. In the sampler 120, the current source 70, the common source input buffer transistor 82, the enable transistor 88 and the steering transistor 90 of FIG. 3A are represented by a first input buffer assembly 122A. A substantially-identical second input buffer assembly 122B is arranged with the first input buffer assembly to define a differential input port 124.
[0055] The charge-transfer transistor 86, sampling capacitor 50 and output capacitor 48 of FIG. 3A are represented by first and second output assemblies 126A and 126B that differentially drive a differential amplifier 128 which forms a differential output port 130. Each input buffer transistor 82 is coupled to a respective one of the sampling capacitors 50. The sources of the charge-transfer transistors 86 are coupled to the common-mode signal Scmand their gates are coupled to receive the transfer signal Strns.
[0056] A single charging transistor 84 is coupled across the differential input of the differential amplifier 128 and responds to the charging signal SCHRG by substantially shorting together the top plates of the sampling capacitors 50 during the sampling time periods. Bias transistors 132 and 133 may be respectively coupled to the inverting and noninverting ports of the differential amplifier 128 to establish suitable reference voltages Vrefand Vrefduring the sampling time periods when the charging transistor 84 is on.
[0057] Small resistors 134 (e.g., on the order of 100 ohms) may optionally be inserted between the input buffer assemblies 122A and 122B and the sampling capacitors Csto further limit parasitic coupling (i.e., suppress signal feedthrough) to the sampling capacitors 50. Operation of the sampler 120 is functionally equivalent to that of the sampler 80 of FIG. 3A except that a differential input signal Sinis differentially processed to realize a differential output voltage Voutat the differential output port 130.
[0058] FIG. 4B illustrates a differential BiCMOS embodiment 140 of the sampler 100 of FIG. 3B. The first and second output assemblies 126A and 126B, the differential amplifier 128, the differential output port 130, resistors 132, the charging transistor 84 and the bias transistors 132 and 133 of the sampler 100 are repeated in the sampler 100.
[0059] The first and second input buffer assemblies 122A and 122B of the sampler 100 are, however, replaced with first and second input buffer assemblies 142A and 142B that are arranged to form the differential input port 124. Each of these latter assemblies include the current source 70, the common emitter input buffer transistor 102 and the charge-transfer transistor 106 of FIG. 3B.
[0060] To insure that the same common-mode signal Scmis established on the bottom plates of the sampling capacitors 50 during the holding time periods, a shorting transistor 144 is preferably arranged across the bottom plates to short them together in response to the transfer signal Strns. Operation of the sampler 140 is functionally equivalent to that of the sampler 100 of FIG. 4A except that a differential input signal Sinis differentially processed to realize a differential output voltage Voutat the differential output port 130.
[0061] Simulations of the samplers of the invention have indicated significant performance improvements at high sampling rates (e.g:, >100 megasamples/second) and with large sampling capacitors (e.g., >500 femtofarads). Samplers of the invention are therefore especially suited for use in ADCs. For example, FIG. 5 illustrates an exemplary subranging ADC 160 that receives analog input signals Sinat an input port 161 and converts them to digital signals at an output port 190 with the aid of samplers 162, 174 and 186. Sampler 162, for example, can be advantageously realized with the methods and structures of the present invention.
[0062] From the sampler 162, samples of the input signal Sinat the input port 161 are converted in an initial ADC 164 to an initial set of digital bits which are delivered to a digital processor 166. An initial digital-to-analog converter (DAC) 168 converts the initial set of digital bits to a converted analog signal which is subtracted from the sampled input signal in a differencer 170 to form an initial residue signal 172. Because this action results in an amplitude reduction, the initial residue signal is preferably “gained up” in an amplifier 173 and then sampled in the subsequent sampler 174.
[0063] The initial conversion process is then repeated. That is, the sampled residue signal is converted in a subsequent ADC 176 to a subsequent set of digital bits which are delivered to the digital processor 166. A subsequent DAC 178 then converts the subsequent set of digital bits to a converted analog signal which is subtracted from the sampled residue signal in a differencer 180 to form a subsequent residue signal 182.
[0064] The subsequent residue signal passes through another amplifier 184 and is sampled in a final sampler 186. A final ADC 188 converts the sampled subsequent residue signal into a final set of digital bits which are combined in the digital processor 166 with the initial and subsequent sets of digital bits to form the final digital output signal at the output port 190.
[0065] To facilitate their description, the plates of a sampling capacitor (e.g., plates 63 and 76 in FIG. 2) have been referred to respectively as bottom and top plates. More generally, they may be termed input and output plates or, more simply, first and second plates.
[0066] It is noted that the noun “charge” has been used herein as it is conventionally defined, i.e., electrical energy that is stored, for example, in a capacitor and also the quantity of the stored electrical energy. It is further noted that the verb “charging” has been used herein as it is conventionally defined, i.e., as the process of storing electrical energy in a capacitor).
[0067] Apparatus of the invention for sampling an input signal Sinand holding a corresponding output voltage Voutthrough successive and alternate sampling and holding time periods has been generically referred to herein as samplers (e.g., the sampler 40 of FIG. 3). It is noted, however, that a variety of equivalent terms are generally used (e.g., sample and hold, sample-and-hold amplifier (SHA) and track and hold circuit).
[0068] The preferred embodiments of the invention described herein are exemplary and numerous modifications, variations and rearrangements can be readily envisioned to achieve substantially equivalent results, all of which are intended to be embraced within the spirit and scope of the invention as defined in the appended claims.
Claims
We claim:
1. A method of sampling an input signal Sinand holding a corresponding output voltage Voutthrough successive and alternate sampling and holding time periods wherein said input signal Sinhas a common-mode signal Scmcomponent, the method comprising the steps of:
during said sampling time periods,
a) enabling an input buffer that receives said input signal Sin; and
b) charging a sampling capacitor Csthrough said input buffer to a charge that corresponds to said input signal Sin;
and during said holding time periods,
a) disabling said input buffer to thereby isolate it from said sampling capacitor Cs; and
b) coupling said sampling capacitor Csbetween said common-mode signal Scmand a sampler output port to provide said output voltage Vout.
2. The method of claim 1, wherein:
said enabling step includes the step of steering a supply current to said input buffer; and
said disabling step includes the step of steering said supply current away from said input buffer.
3. The method of claim 1, wherein said coupling step includes the step of transferring electrical charge from said sampling capacitor Csto an output capacitor Co.
4. The method of claim 3, wherein said transferring step includes the steps of:
connecting said output capacitor Cobetween an inverting input port and an output port of a differential amplifier; and
connecting said sampling capacitor Csbetween said common-mode signal Scmand said inverting input port.
5. The method of claim 3, further including the step of discharging said output capacitor Coduring said sampling time periods.
6. A sampler for sampling an input signal Sinand holding a corresponding output voltage Voutthrough successive and alternate sampling and holding time periods wherein said input signal Sinhas a common-mode signal Scmcomponent, the sampler comprising:
an input buffer that is disabled during said holding periods and that is enabled during said sampling periods to provide a version of said input signal Sinto a buffer output port of said input buffer;
a sampling capacitor Cshaving a first plate that is coupled to said buffer output port and a second plate that is coupled to a sampler output port;
a charging switch Scwhich enables current flow through said sampling capacitor Csduring said sampling time periods; and
a charge-transfer switch Sc-twhich is open during said sampling time periods and which couples said first plate to said common-mode signal Scmduring said holding time periods;
said sampling capacitor Csthereby charged during said sampling time periods to an electrical charge that corresponds to said input signal Sinwith said second plate providing said output voltage Voutto said sampler output port during said holding time periods.
7. The sampler of claim 6, further including:
a differential amplifier having an inverting input port coupled to said second plate, a noninverting input port and an amplifier output port that forms said sampler output port; and
an output capacitor Cothat is connected between said inverting input port and said amplifier output port.
8. The sampler of claim 6, further including a current-source circuit that provides a supply current to said input buffer to enable said input buffer during said sampling time periods and removes said supply current from said input buffer to disable said input buffer during said holding time periods.
9. The sampler of claim 8, wherein said current-source circuit includes:
a current source which generates said supply current; and
a current switch which couples said supply current to said input buffer during said sampling time periods and removes said supply current from said input buffer during said holding time periods.
10. The sampler of claim 8, wherein said current-source circuit includes:
a current source which generates said supply current; and
a differential pair of first and second transistors coupled to steer said supply current to said input buffer during said sampling time periods and steer said supply current away from said input buffer during said holding time periods.
11. The sampler of claim 10, wherein said charging switch Sc, said charge-transfer switch Sc-tand said first and second transistors are metal-oxide semiconductor transistors.
12. The sampler of claim 8, wherein said current-source circuit includes a current source that provides said supply current to said input buffer during said sampling time periods and wherein said charge-transfer switch Sc-tis arranged to steer said supply current away from said input buffer during said holding time periods.
13. The sampler of claim 6, wherein said input buffer is a metal-oxide semiconductor transistor whose gate is coupled to receive said input signal Sinand whose source forms said buffer output port.
14. The sampler of claim 6, wherein said input buffer is a first bipolar junction transistor whose base is coupled to receive said input signal Sinand whose emitter forms said buffer output port and said charge-transfer switch Sc-tis a second bipolar junction semiconductor transistor that is coupled to form a differential pair with said first bipolar junction semiconductor transistor.
15. The sampler of claim 6, further including a discharge switch Sdarranged to discharge said output capacitor Coduring said sampling time periods.
16. An ananalog-to-digital converter system that converts an input signal Sinhaving a common-mode signal Scmcomponent to a corresponding digital output signal, comprising:
a sampler that is arranged to sample said input signal Sinand hold a corresponding output voltage Voutthrough successive and alternate sampling and holding time periods, said sampler including:
a) an input buffer that is disabled during said holding periods and that is enabled during said sampling periods to provide a version of said input signal Sinto a buffer output port of said input buffer;
b) a sampling capacitor Cshaving a first plate that is coupled to said buffer output port and a second plate that is coupled to a sampler output port;
c) a charging switch Scwhich enables current flow through said sampling capacitor Csduring said sampling time periods; and
d) a charge-transfer switch Sc-twhich is open during said sampling time periods and which couples said first plate to said common-mode signal Scmduring said holding time periods to provide said output voltage Voutat said second plate; and
a analog-to-digital converter that converts said output voltage Voutto a respective set of digital bits of said digital output signal.
17. The system of claim 16, wherein said sampler further includes a current-source circuit that provides a supply current to said input buffer to enable said input buffer during said sampling time periods and removes said supply current from said input buffer to disable said input buffer during said holding time periods.
18. The system of claim 16, wherein said sampler further includes:
a differential amplifier having an inverting input port coupled to said second plate, a noninverting input port and an amplifier output port that forms said system output port; and
an output capacitor Cothat is connected between said inverting input port and said amplifier output port.
19. The system of claim 17, wherein said sampler further includes a discharge switch Sdarranged to discharge said output capacitor Coduring said sampling time periods.
20. The system of claim 16, wherein said analog-to-digital converter includes:
an initial analog-to-digital converter that converts said output voltage Voutto an initial set of digital bits of said digital output signal;
a digital-to-analog converter that converts said initial set of digital bits to an analog signal that is subtracted from said input signal Sinto form an analog residue signal; and
an final analog-to-digital converter that converts said analog residue signal to a final set of digital bits of said digital output signal.