System and method for extending input range of analog-to-digital converter
By using DAC weight preset digital/analog converter during the sampling phase of the analog-to-digital converter, adjusting the digital control input to expand the input range, the input range attenuation problem caused by parasitic capacitance of conventional CR-SAR ADCs is solved, and the signal is improved Noise and distortion ratio and significant digit number.
Patent Information
- Application Number
- CN202410696502.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-06-02
- Filing Date
- 2024-05-31
- Publication Date
- 2024-12-03
AI Technical Summary
Conventional top plate sampling charge redistribution successive approximation register (CR-SAR) analog-to-digital converter (ADC) causes input range attenuation due to parasitic capacitance on the top capacitor plate, affecting signal-to-noise and distortion ratio (SNDR) and significant bit count (ENOB ).
By using available DAC weight preset digital/analog converters during the sampling phase, adjusting the digital control input to expand the input range of the analog/digital converters, presetting the digital control input with preset values to ensure that the input voltage is within the full range. And determine the digital output by successive approximation register conversion.
Effectively extends the input range of the analog-to-digital converter, improves signal-to-noise and distortion ratio (SNDR), reduces the number of effective bits (ENOB) losses, and ensures the integrity and accuracy of the input signal.
Smart Images

Figure CN119070823A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates generally to analog-to-digital conversion, and more particularly to a system and method for extending the input range of a digital-to-analog converter (DAC) by presetting the DAC using available DAC weights during a sampling phase to increase the input range. Background Art
[0002] A conventionally operated top-plate sampled charge redistribution successive approximation register (CR-SAR) analog-to-digital converter (ADC) has the disadvantage of having its input range attenuated by parasitic capacitance on the top capacitive plate (as well as other conversion issues), which has a negative impact on the SNDR (signal-to-noise and distortion ratio) and the corresponding effective number of bits (ENOB). Summary of the invention
[0003] According to one embodiment, an analog / digital converter includes:
[0004] a digital-to-analog converter (DAC) including a sampling node coupled to a capacitor array that receives a digital control input;
[0005] a comparator coupled to compare the voltage of the sampling node with a reference voltage to generate an input comparison value; and
[0006] A controller is coupled to preset the digital control input with a preset value, prompt the DAC to sample an input voltage onto the sampling node, reset the digital control input, and then perform conversions to determine a digital output by continuously adjusting the digital control input based on the input compare value.
[0007] In one or more embodiments, the analog-to-digital converter further includes an adder coupled to subtract a digital adjustment value indicative of the preset value from the digital output to provide an adjusted digital output.
[0008] In one or more implementations, the analog-to-digital converter further includes an input switch coupled to sample the input voltage onto the sampling node.
[0009] In one or more embodiments, the analog / digital converter further comprises:
[0010] a sample predictor coupled to provide a predicted sample value based on the input voltage; and
[0011] Wherein the controller is configured to select the preset value based on the predicted sample value.
[0012] In one or more embodiments, the DAC has a full-scale range between an upper reference voltage level and a lower reference voltage level, and wherein when the predicted sample indicates that the input voltage is outside the full-scale range, the controller is configured to select the preset value so that the voltage of the sampling node is adjusted to a voltage within the full-scale range when the digital control input is reset.
[0013] In one or more embodiments, the DAC comprises a top plate sampling DAC.
[0014] In one or more embodiments, the capacitor array includes a plurality of capacitors, each capacitor having a first terminal coupled to the sampling node and each capacitor having a second terminal coupled to an output of a corresponding one of a plurality of drivers, wherein each of the plurality of drivers has an input receiving a corresponding one of a plurality of bits of the digital control input, and wherein each of the plurality of drivers is coupled to drive a corresponding one of the plurality of capacitors to a selected differential reference voltage of a first and a second differential reference voltage based on a corresponding one of the plurality of bits.
[0015] In one or more embodiments, the DAC has a full-scale range between an upper reference voltage level and a lower reference voltage level, and wherein the controller is configured to select the preset value so that the voltage of the sampling node is adjusted to a voltage between the upper reference voltage level and the lower reference voltage level when the digital control input is reset.
[0016] In one or more implementations, the converting includes a successive approximation register conversion.
[0017] In one or more embodiments, the controller includes: a range extension controller coupled to preset the digital control input with a preset value, prompt the DAC to sample the input voltage onto the sampling node, reset the digital control input, and then prompt the conversion; and a successive approximation register controller coupled to perform the conversion to determine the digital output by continuously adjusting the digital control input based on the input comparison value when the prompt is received.
[0018] According to another embodiment, a method of converting an analog input voltage into a digital value includes:
[0019] Presetting a digital control input of a digital / analog converter (DAC) with a preset value;
[0020] Sampling the input voltage to a sampling node of the DAC;
[0021] resetting the digital control input of the DAC;
[0022] comparing the voltage of the sampling node with a reference voltage to generate an input comparison value;
[0023] as well as
[0024] Conversion is performed by continuously adjusting the digital control input based on the input comparison value to determine a digital output.
[0025] In one or more embodiments, the method further includes subtracting a digital adjustment value indicative of the preset value from the digital output to provide an adjusted digital output.
[0026] In one or more embodiments, the sampling includes prompting for input of a sampling switch.
[0027] In one or more embodiments, the method further comprises:
[0028] Predict the next sample value; and
[0029] The preset value is determined based on the predicted sample value.
[0030] In one or more embodiments, determining the preset value includes determining the preset value so that the voltage of the sampling node is adjusted to a voltage within a full-scale range of the DAC when the digital control input of the DAC is reset.
[0031] In one or more embodiments, determining the preset value includes determining the preset value so that the voltage of the sampling node is adjusted as closely as possible to a center voltage of a full-scale range of the DAC when the digital control input of the DAC is reset.
[0032] In one or more embodiments, the method further includes configuring the DAC as a top plate sampling DAC.
[0033] In one or more embodiments, the method further includes: configuring the DAC as a plurality of capacitors, each capacitor having a first terminal coupled to the sampling node and each capacitor having a second terminal coupled to an output of a corresponding one of a plurality of drivers; providing each of the plurality of drivers with an input that receives a corresponding one of a plurality of bits of the digital control input; and configuring each of the plurality of drivers to drive a corresponding one of the plurality of capacitors to a selected differential reference voltage of a first and a second differential reference voltage based on a corresponding one of the plurality of bits.
[0034] In one or more embodiments, the method further includes performing the presetting, sampling, resetting, comparing, and converting using a charge redistribution successive approximation register analog-to-digital converter.
[0035] In one or more implementations, performing the conversion includes performing a successive approximation register conversion. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Embodiments of the present invention are illustrated by way of example in the accompanying drawings, and the embodiments of the present invention are not limited by the accompanying drawings. Similar reference numerals in the drawings may indicate similar elements. For simplicity and clarity, the elements in the drawings are shown, and these elements are not necessarily drawn to scale.
[0037] Figure 1 is a simplified block diagram of a differential CR-SAR ADC configured with input range extension according to one embodiment.
[0038] Figure 2 is configured as a top sampling differential split capacitor DAC according to one embodiment Figure 1 Schematic diagram of the DAC.
[0039] Figure 3 is a graphical diagram plotting FSR range for a conventional configuration, an “ideal” configuration, and a compensated FSR configuration with range extension as further described herein, according to one embodiment.
[0040] Figure 4 is a simplified block diagram of a differential CR-SAR ADC configured with input range extension according to one embodiment.
[0041] Figure 5 is a visualization diagram according to one embodiment, which shows a 1000-bit 1000 having the first three most significant bits (MSBs). Figure 4 A simplified depiction of the five operating stages of the ADC sampling DAC and the curve of the voltage VS at the sampling node.
[0042] Figure 6 According to an embodiment of the present invention, Figure 4 Flowchart of the operation of the differential CR-SAR ADC.
[0043] Figure 7 According to one embodiment, Figure 4 The range extension controller is Figure 6 A flow chart of the operations for determining a digital offset during operation is shown in FIG.
[0044] Figure 8 is a graphical diagram plotting an exemplary VIN signal, a voltage VS at a sampling node of a differential CR-SAR ADC, and a maximum range of a top sampling DAC of the ADC, according to one embodiment. DETAILED DESCRIPTION
[0045] A charge redistribution successive approximation register (CR-SAR) analog-to-digital converter (ADC) incorporates a top plate sampling digital-to-analog converter (DAC) that samples an analog input voltage during each of a series of sampling cycles. In each sampling cycle, the ADC predicts the next input sample and accordingly "presets" the sampling DAC by adjusting a digital control value during the input sampling period to effectively add a digital offset. The digital control value of the DAC is then reset back to its normal initial value, and a normal SAR conversion is performed to determine a first digital output value. A digital preset value corresponding to the digital offset is subtracted from the first digital output value to provide a final or adjusted digital output value.
[0046] Figure 1 1 is a simplified block diagram of a differential CR-SAR ADC 100 configured with input range extension according to one embodiment. A differential analog input voltage (VIN) includes a positive polarity VIN+ and a negative polarity VIN- provided to respective inputs of a pair of sampling switches 102 and 104. In response to a sampling control signal SMP, VIN+ is sampled by a first digital-to-analog converter (DAC) 106 that produces a positive sample voltage VS+, and VIN- is sampled by a second DAC 108 that produces a negative sample voltage VS-. VS+ and VS- are provided to respective inputs of a comparator 110 having an output that provides an input comparison value CMP to an input of a SAR and range extension controller 112. The SAR and range extension controller 112 asserts the SMP signal to periodically sample VIN (e.g., every sampling clock cycle, etc.) and then performs successive approximation based on the CMP value. Specifically, the SAR and range extension controller 112 updates a set of positive digital control values DP[i] provided to the input of the DAC 106 and updates another set of negative digital control values DN[i] provided to the input of the DAC 108 based on the CMP value. In this way, the SAR and range extension controller 112 performs a conversion of VIN to output a corresponding digital output value DOUT.
[0047] In conventional operation of the differential CR-SAR ADC 100 without range extension, the SAR and range extension controller 112 initially asserts DP[i] and DN[i] to reset the values, thereby placing the DACs 106 and 108 in an initial reset state. The SAR and range extension controller 112 then toggles SMP so that VIN is sampled by the DACs 106 and 108 to generate initial values for the VS+ and VS- sampled voltages. The SAR and range extension controller 112 then performs a successive approximation register (SAR) conversion by adjusting DP[i] and DN[i] one bit at a time while monitoring CMP to ultimately produce a digital value that is provided to or otherwise converted to DOUT. However, conventional operation of the ADC 100 suffers from input range attenuation due to parasitic capacitance on the top plate of each of the DACs 106 and 108, which reduces the effective number of bits (ENOB). When the input voltage is outside the attenuated range, the input voltage may be scaled prior to SAR conversion, which has a negative impact on the signal-to-noise and distortion ratio (SNDR). For example, in one embodiment, normal operation results in a 10-30% full scale range (FSR) loss.
[0048] The SAR and range extension controller 112 is modified relative to the conventional configuration to implement input range extension to extend the operating range beyond the standard FSR. When operating with range extension as further described herein, instead of initially resetting the DACs 106 and 108 for sampling, the SAR and range extension controller 112 initially presets both the DACs 106 and 108 by modifying both DP[i] and DN[i] with digital preset values. When both the DACs 106 and 108 are pre-filled with the preset values, the SAR and range extension controller 112 asserts SMP to sample VIN into the DACs 106 and 108. Next, the SAR and range extension controller 112 "resets" the DACs 106 and 108 by returning both DP[i] and DN[i] to their reset values while both remain filled with corresponding VIN samples. Next, the SAR and range extension controller 112 performs a SAR conversion process to generate a "preliminary" output value PDOUT, and then subtracts the corresponding preset value from PDOUT to generate an adjusted or final digital output DOUT.
[0049] The initial preset of the DP[i] and DN[i] values effectively adds a digital offset or preset value to the DACs 106 and 108 during sampling. The magnitude and direction of the preset value may be based on a prediction of the input sample, as further described herein. When the magnitude of the analog input signal is greater than the standard FSR of the ADC, the preset is added to achieve sampling within the operating range of the DAC. It should be appreciated that the analog input signal is not attenuated prior to conversion, thereby maintaining input signal integrity. After conversion, the preset value is subtracted to provide an accurate digital representation of the input sample. It should be noted that the initial "addition" of the digital offset may be implemented as adding a positive or negative value, where the digital offset value is then subtracted from the first digital value to provide an adjusted digital output value.
[0050] Figure 2 1 is a schematic diagram of a DAC 106 configured as a top sampling differential split capacitor DAC according to a conventional configuration. The sampling switch 102 receives VIN+, and one terminal of the sampling switch is coupled to a sampling node 202 of the DAC 106 which produces a sampled voltage VS+. The positive digital value DP[i] is subdivided into N+1 top signals DP_TOP[0], DP_TOP[1], ..., DP_TOP[N] and N+1 bottom signals DP_BOT[0], DP_BOT[1], ..., DP_BOT[N], where N is any suitable positive integer for achieving a desired DAC resolution. The top signals DP_TOP[0..N] are provided to respective inputs of N+1 inverting drivers 204a, 204b, ..., 204c, the outputs of which are coupled to respective terminals (or plates) of N+1 capacitors 206a, 206b, ..., 206c. In a similar manner, the bottom signal DP_BOT[0..N] is provided to respective inputs of N+1 inverting drivers 208a, 208b, ..., 208c, the outputs of which are coupled to respective terminals of another set of N+1 capacitors 210a, 210b, ..., 210c. The opposite terminal of each of the capacitors 206a-206c and 210a-210c is coupled to the sampling node 202. Each of the inverting drivers 204a-204c and 210a-210c switches between the polarity of the differential reference voltage VrefP and VrefN.
[0051] In the illustrated configuration, capacitors 206a-206c and capacitors 210a-210b each have a capacitance based on capacitance C, where the capacitor array forms a binary capacitance distribution of C / 2, C / 4, ..., C / 2. N+1, from the most significant bit numbered "0" to the least significant bit numbered "N". Parasitic capacitance Cp is shown coupled between sampling node 202 and VrefN. It should be noted that the illustrated binary distribution is merely exemplary, and that many alternative distribution configurations, including non-binary distributions, are possible and are contemplated. Although not explicitly shown, DAC 108 may be configured in substantially the same manner as being controlled by a negative digital value DN[i] for generating a negative polarity sample voltage VS-, and is not further described. In addition, Figure 2 The DAC configuration shown in is exemplary only, where it should be appreciated that many different alternative top sampling DAC configurations may be used.
[0052] It will be appreciated that one may, for example, Figure 2 DACs 106 and 108 are configured in the conventional manner shown in , but DACs 106 and 108 operate in a new manner as described herein to achieve overall range extension. Each of DACs 106 and 108 may have a reset state. Figure 2 , for example, the top signals DP_TOP[0..N] may each be driven to a logic "0" (voltage at or near VrefN) such that the upper capacitors 206a-206c are driven to VrefP, and the bottom signals DP_BOT[0..N] may each be driven to a logic "1" (voltage at or near VrefP) such that the lower capacitors 210a-210c are driven to VrefN. Instead of sampling when the DACs 106 and 108 are in their reset states according to conventional configurations, one or more up to all of the DP_TOP[i] or DN_TOP[i] signals may instead be driven to a logic "1", or one or more up to all of the DP_BOT[i] or DN_BOT[i] signals may instead be driven to a logic "0" to place the corresponding DACs in a preset state for sampling. After sampling, the DP[i] and DN[i] control signals are reset back to their initial values to "reset" the DACs 106 and 108 while filling with the sampled input values. The SAR and range extension controller 112 then performs a normal SAR conversion by adjusting DP[i] and DN[i] one bit at a time while monitoring CMP to ultimately produce a digital value PDOUT. The preset value added during sampling is then subtracted from PDOUT to provide DOUT as the adjusted output value. The modified operation overcomes the deficiencies of the conventional operation of the differential CR-SAR ADC 100 to achieve considerable range extension.
[0053] Figure 3is a graphical diagram that plots the FSR range for a conventional configuration, an "ideal" configuration, and a compensated FSR configuration with range extension as further described herein, according to one embodiment. The input range is centered about a center voltage shown as CNTR. The ideal FSR range is between a higher voltage IDEAL FSR+ and a lower voltage IDEAL FSR-, with an overall range of FSR IDEAL , which may be the same range as the reference voltages VrefP and VrefN, respectively. For a differential configuration, CNTR may be 0. In a single-ended configuration, IDEAL FSR- may be zero (or some other convenient reference voltage level), and CNTR is FSR IDEAL / 2.
[0054] The conventional configuration with top plate sampling is limited to a reduced voltage range between the higher voltage CONV FSR+ and the lower voltage CONV FSR-. Overall FSR FSR of the conventional configuration CONV It can be expressed by the following equation (1):
[0055]
[0056] Among them C DACT is the total capacitance of the DAC. IDEAL =1(Equation (1), where C P =0), the parasitic capacitance C P The presence of reduces the overall range of common configurations to less than 1. In many feasible configurations, C P Relatively large, making FSR CONV Usually FSR IDEAL About 70%-90%.
[0057] When operating with range extension as described herein, the operating range extends to the range between the higher voltage COMP FSR+ and the lower voltage COMP FSR-. COMP It can be expressed by the following equation (2):
[0058]
[0059] Among them C DACP is the equivalent preset capacitance value for the DAC. In this way, depending on C DACP The value of FSR COMP Relative to FSR CONV Increase, and may even greatly exceed FSR IDEAL In one embodiment, for example, FSR COMP Relative to FSR IDEALIn conventional configurations of top sampling SAR ADCs, if the analog input voltage rises above the feasible FSR, typically the input voltage is divided before the conversion cycle begins to fit the feasible range of FSR, which reduces the SNR. In a compensated SAR ADC as described herein, the DAC is preset and samples the unmodified input voltage, and the resulting first output digital value is then adjusted by subtracting a preset factor to achieve a compensated digital output value.
[0060] It should be noted that embodiments are described herein for a differential configuration, wherein it should be appreciated that the range extension as described herein is equally applicable to a single-ended configuration. The single-ended CR-SAR ADC may be configured in a similar manner to the differential CR-SAR ADC 100, except that a single DAC is utilized. The lower reference voltage VrefN may be replaced by a single reference voltage level, which is typically ground or zero (0) volts (V). See Figure 3 , the IDEAL FSR- voltage for the differential configuration is VrefN, which is the same magnitude as the IDEAL FSR+ with opposite polarity, while the IDEAL FSR- voltage for the single-ended configuration can be 0 V. The CNTR voltage for the differential configuration can be at or close to 0 V, while the CNTR for the single-ended configuration is FSR / 2. Otherwise, the operational concepts are substantially the same.
[0061] Figure 44 is a simplified block diagram of a differential CR-SAR ADC 400 configured with input range extension according to one embodiment. In one embodiment, the SAR and range extension controller 112 may be separately configured as a SAR controller 406 and a range extension controller 408. An analog input signal VIN is provided to the input of a sampling DAC 402, which provides a sampled voltage to a comparator 404. Based on a comparison of VS with a reference voltage Vref (e.g., VS+ / - with VRefP / N), the comparator 404 provides an input comparison value CMP to the input of the SAR controller 406. The range extension controller 408 initially provides a digital control input signal D to apply a preset value to the sampling DAC 402, then prompts the sampling DAC 402 to sample VIN when the sampling DAC 402 is preset, and then resets the digital control input signal D of the sampling DAC 402. The range extension controller 408 then prompts the SAR controller 406 to perform a SAR conversion process by adjusting the digital control input signal D one bit at a time based on CMP to produce a digital output PDOUT. PDOUT is provided to one input of a digital adder 410, which receives a PRESET value from the range extension controller 408. The PRESET value corresponds to the preset value of the sampling DAC 402. The adder 410 subtracts PRESET from PDOUT to provide an adjusted or final digital output value DOUT. The ADC 400 includes a sample predictor 412 that receives one or both of VIN and DOUT and generates a predicted sampled signal PS that is provided to the range extension controller 408 during each sampling cycle. The range extension controller 408 determines the preset value for presetting the sampling DAC 402 for each sampling cycle. The entire process is repeated in consecutive sampling cycles.
[0062] The sampling DAC 402 may be implemented as a differential DAC including a pair of DACs configured in a similar manner to the DACs 106 and 108, which in turn may be coupled to the DACs 106 and 108. Figure 2 . In the differential configuration, VIN and VS are differential signals, and the digital control signal D can be subdivided into two pairs of top and bottom control signals in a similar manner as previously described. The sampling DAC 402 can also include a pair of sampling switches that are configured to operate in substantially the same manner as the sampling switches 102 and 104 described previously. In the single-ended configuration, VIN and VS are single-ended signals referenced to ground, etc., and the sampling DAC 402 can include a single top sampling DAC.
[0063] The sample predictor 412 may be configured to generate a PS based on the last sample of VIN as indicated by DOUT. In addition, the sample predictor 412 may perform some level of statistical analysis and slope measurement to determine the direction and rate of change of VIN for use in then determining the next PS on a sampling cycle by sampling cycle basis. The range extension controller 408 uses the PS to determine the preset value to apply to the sampling DAC 402 on a sampling cycle by sampling cycle basis. In some embodiments, if VIN is well within the conventional FSR (e.g., at or close to the CNTR ( Figure 3 )), the preset may be zero or no preset is applied. In some embodiments, if VIN is not close enough to the outer range limit of the sampling DAC 402, the preset may be zero or no preset is applied. Alternatively, after sampling VIN, for the sampling DAC 402, the preset value may be determined to cause the sampling DAC 402 to reset as close to the CNTR as possible. In some embodiments, when clipping is detected such that VIN exceeds the normal range of the sampling DAC 402, the range extension controller 408 may apply a larger preset up to the maximum preset in an attempt to cause the sampling DAC 402 to reset completely within the normal FSR after sampling VIN.
[0064] Figure 5 FIG. 4 is a diagrammatic representation of five operation phases of a simplified depiction of a sampling DAC 402 of an ADC 400 with the first three most significant bits (MSBs) and a curve of a voltage VS at a sampling node 502 according to one embodiment. The structure of the sampling DAC 402 may be similar to Figure 2 The structure of the DAC 106 is shown in Figure 2 The inverting driver is not shown in FIG. In fact, the visualization only shows the output voltage as a result of switching the inverting driver during each of the operating phases of the conversion process. Only the first three capacitor stages and the parasitic capacitance C P , where it will be appreciated that additional capacitor stages may be included. The capacitor values are shown as a ratio to the total capacitance (e.g., 1 / 4, 1 / 8, 1 / 16, etc.) rather than as a ratio to the capacitance value C. For ease of illustration, the positive reference VrefP in this example is set to 1V and the negative reference VrefN is set to 0V, where it will be appreciated that alternative reference voltages may be used. The input sampling switch 102 is shown coupled to the sampling node 502 to facilitate sampling of VIN by the sampling DAC 402.
[0065] The operation phase includes a preset phase 510, a reset phase 512, and a series of phases illustrating a normal SAR conversion process for determining three MSB bits, including an MSB phase 514, an MSB-1 phase 516, and an MSB-2 phase 518. The reset phase 512 illustrates that the control voltage is in a reset state of the sampling DAC 402, where the top control voltages are each 1V and the bottom control voltage is 0V. In the first preset phase 510, the range extension controller 408 switches the top control voltages of the MSB and MSB-1 from 1V (normal reset state) to 0V to apply the preset value. The curve of VS shows a normal FSR between CONVFSR- and CONVFSR+, where VS is less than CONV FSR- but greater than COMP FSR-. The sample predictor 412 estimates the voltage level of VS and outputs PS, and the range extension controller 408 uses PS to determine the preset value V_PRE, which in this case means switching both the MSB and MSB-1 bits down as shown. This is considered a “positive” preset value because when reset after sampling VIN, the voltage of VS rises by V_PRE, as shown in reset phase 512 .
[0066] It should be noted that alternative preset values are carefully considered in different situations and under different operating conditions. Instead of pushing the MSB and MSB-1 down from 1V to 0V as shown, any number of MSBs may be switched, such as only one MSB, both MSB and MSB-1, MSB and MSB-1 and MSB-2, and so on, through all top bits. When a negative preset value is applied in the opposite direction from bottom to top in the preset phase instead, then any number of bottom voltages may be switched from 0V to 1V, such as MSB, MSB and MSB-1, MSB and MSB-1 and MSB-2, and so on, through all bottom bits. In other words, the preset value may have any magnitude based on the resolution or number of DAC bits, and may be in either direction, upward or downward.
[0067] In the preset phase 510, after applying the preset value, the sampling switch 102 is toggled to sample VIN into the sampling node 502, which drives VS to have the same voltage level as VIN. The operation then proceeds to the reset phase 512, in which the range extension controller 408 opens the sampling switch 102 to allow the sampling node 502 to float, and then switches the digital control value and the corresponding control voltage back to the reset condition. In the reset condition as illustrated in the reset phase 512, the top control voltages all return to 1V, and the bottom control voltage remains at 0V. As shown by the curve of VS, this causes the voltage of the sampling node 502 to increase by V_PRE to a voltage level VIN+V_PRE, which is well within the conventional range of the DAC, meaning higher than CONVFSR- but less than CONVFSR+. In this way, the preset voltage V_PRE has repositioned the voltage level of VS to enable the DAC to perform conventional SAR conversions.
[0068] Once the voltage of VS settles during the reset phase 512, the range extension controller 408 prompts the SAR controller 412 to perform a SAR conversion while the sampling switch 102 remains open. In the next series of phases, the SAR controller 412 determines the DAC bit one bit at a time, starting with the MSB in the MSB phase 514. In each conversion phase, the voltage of VS is compared to the SAR conversion point (SCP), which may be a differential zero (e.g., the center point voltage between VrefP and VrefN). As shown, the SAR controller 412 detects that VS is greater than SCP and switches the top control voltage of MSB from 1V to 0V, causing VS to decrease accordingly. The CMP value indicates that VS is still above SCP, so that in the next phase MSB-1 516, the SAR controller 412 switches the top control voltage of MSB-1 from 1V to 0V, causing VS to decrease accordingly. The CMP value indicates that VS has fallen below SCP, so that in the next stage MSB-2 518, the SAR controller 412 switches the bottom control voltage of MSB-2 from 0V to 1V, causing VS to decrease accordingly. The operation continues for each sequential bit in the sequential cycle. In a simplified illustration, the operation is completed and VS is at or substantially close to SCP.
[0069] Although not shown, the digital value represented by the control bit corresponds to the digital output PDOUT. Figure 4 , digital adder 410 subtracts a PRESET value (which is equal to or otherwise corresponds to the digital value applied during preset phase 510 corresponding to voltage V_PRE) from PDOUT to determine an adjusted digital output value DOUT. The operation is repeated in a continuous loop to convert VIN to a series of digital values on DOUT.
[0070] Figure 6 6 is a flow chart outlining the operation of the differential CR-SAR ADC 400 according to one embodiment. At a first block 602, the next sample PS is predicted by the sample predictor 412 based on VIN, DOUT, and any other selected parameters and algorithms. Various methods or algorithms for predicting the next sample are possible, which may include estimating the slope and direction of VIN and making a prediction based on the last sampled value of VIN. Operation proceeds to block 604, where the range extension controller 408 determines and generates a digital preset value based on PS. The digital preset value includes both magnitude and direction to attempt to adjust the voltage of the sampling node to be at or close to CNTR when the DAC is reset after sampling. At the next block 606, the range extension controller 408 applies the digital preset value to the DAC (e.g., a digital control input signal D that presets the sampling DAC 402). At the next block 608, the range extension controller 408 closes one or more sampling switches (e.g., 102, 104, etc.) to sample VIN onto the DACs (e.g., to the corresponding sampling nodes of DACs 106 and 108 (e.g., 106, 108) while the DACs are still in the preset condition. Figure 2 402). At the next block 610, the range extension controller 408 resets the digital input control signal of the DAC, which effectively applies a preset to place the sampled voltage within the conversion range of the DAC. At the next block 612, the range extension controller 408 prompts the SAR controller 406 to perform the SAR conversion process one bit at a time from MSB to LSB to generate a digital output value PDOUT. At the next block 614, the digital adder 410 subtracts the digital preset value from PDOUT to provide DOUT, and the current conversion cycle is completed. The operation proceeds to block 616 to ask whether to continue the conversion process. If so, the operation returns back to block 602, and the conversion process is performed again for the next sample of VIN. The operation can continue in this manner to convert VIN to a series of digital values on DOUT.
[0071] Figure 76 is a flow chart showing the operation of the range extension controller 408 to determine a digital preset value (or digital offset) at block 604, according to one embodiment. At a first block 702, the range extension controller 408 receives a predicted sample PS from the sample predictor 412. Depending on the magnitude of the PS, the operation proceeds to one of several blocks shown as blocks 704, 710, and 720. When the PS indicates a minimum (MIN) range sample at or near the CNTR (which is the center of the FSR as determined at block 704), then the operation proceeds to block 706 where the digital preset value is not applied or is effectively zero. In this case, the sampling DAC is simply set to its reset condition in the preset phase 510 when VIN is sampled, and the control voltage remains unmodified during the reset phase 512 when the sampling switch is open. Thereafter, the operation is the same. The operation then loops back to block 702 for the next sampling cycle.
[0072] When the PS indicates a mid-range (MID) sample (e.g., as determined at block 710, close to but not exceeding FSR), then the operation proceeds to block 712 to determine the direction of the preset. At block 712, it is inquired whether the sample is near the positive limit or upper limit FSR+ of the DAC range. If so, the operation proceeds to block 714, in which the range extension controller 408 generates and provides a negative MID range preset. In this case, the digital preset value can be a negative value, so that when the DAC transitions from the preset to its reset mode, the sampled voltage VS of the sampling node decreases at or toward CNTR of the DAC voltage range. Otherwise, the sample is near the lower limit FSR- and the operation proceeds to block 716, in which the range extension controller 408 generates and provides a positive MID range preset, so that when the DAC transitions from the preset to its reset mode, the sampled voltage VS of the sampling node increases at or toward CNTR of the DAC voltage range. After block 714 or 716, the operation loops back to block 702 for the next sampling cycle.
[0073] When the PS indicates a maximum (MAX) range sample between or near the maximum limit or even outside the FSR range (as determined at block 720), then the operation proceeds to block 722 to determine the direction of the preset. At block 722, it is queried whether the sample is at or above the positive or upper limit FSR+ of the DAC range. If so, the operation proceeds to block 724, in which the range extension controller 408 generates and provides a negative MAX range preset. In this case, the digital preset value can be a negative value, so that when the DAC transitions from the preset to its reset mode, the sampled voltage VS of the sampling node decreases at or toward CNTR of the DAC voltage range. Otherwise, the sample is at or beyond the lower limit FSR- and the operation proceeds to block 726, in which the range extension controller 408 generates and provides a positive MAX range preset, so that when the DAC transitions from the preset to its reset mode, the sampled voltage VS of the sampling node increases at or toward CNTR of the DAC voltage range. After either block 724 or 726, operation loops back to block 702 for the next sampling cycle.
[0074] Many variations are possible, and these are carefully considered. In an alternative embodiment, a preset value is calculated in each sampling cycle to offset the predicted value to be as close to the center value CNTR as possible given the resolution of the DAC. In a slight variation, when the PS is at or very close to CNTR (e.g., within an empirically determined minimum threshold), no preset is applied, or a predetermined minimum preset may be used in the direction indicated by the PS. In another alternative embodiment, additional ranges may be defined, and a corresponding preset is determined for each defined range, wherein the preset is used to offset the initial value back to or sufficiently close to CNTR after the DAC is reset. In yet another alternative embodiment, the preset is not used unless and until the PS indicates that VIN is in the FSR or extends outside the FSR (above or below FSR-), in which case the PS indicates that VIN is in the FSR or extends outside the FSR, and the maximum preset is applied in the appropriate direction to ensure that the voltage of the sampling node is within the FSR after the DAC is reset.
[0075] In any of the embodiments, when clipping is detected such that VIN is outside the FSR of the DAC, a maximum digital offset may be applied when sampling VIN to preset the DAC so that the voltage of the sampling node is within the FSR of the DAC after the DAC is reset. In this way, VIN remains unmodified so that the SNDR and corresponding effective number of bits (ENOB) remain unmodified.
[0076] Figure 8806 is a diagrammatic representation of an exemplary VIN signal, a voltage VS at a sampling node of a differential CR-SAR ADC (e.g., 400), and a maximum value MAX indicating the maximum positive range of the top sampling DAC of the ADC. As shown at 802, VIN is above MAX, which means that the DAC will not be able to converge properly during normal SAR conversion. In a conventional configuration, the input voltage will first be scaled down, resulting in a loss of SNDR and a reduction in ENOB. In practice, the ADC first presets the DAC with a negative digital offset, and then samples VIN at 802 while it is preset. In the next clock cycle shown at 804, the DAC is reset, which causes VS to decrease to be well within the operating range of the DAC. In the next sequence of clock cycles shown at 806, the ADC performs a standard SAR conversion to determine the DAC bit from MSB to LSB. The ADC determines the final value shown at 808, which is used to generate PDOUT. Although not shown, the ADC subtracts the preset value from PDOUT to determine the final adjusted digital output value DOUT. The operation is repeated in a continuous sampling cycle.
[0077] Although the present invention has been described in conjunction with several embodiments, the present invention is not intended to be limited to the specific forms set forth herein. Rather, it is desired to cover alternatives, modifications, and equivalents that may reasonably be included within the scope of the present invention as defined by the appended claims. For example, in various embodiments in which the present invention is not limited to a particular circuit system polarity, device type, or voltage or error level, etc., variations of positive circuit systems or negative circuit systems may be used. For example, circuit system states such as circuit system low and circuit system high may be reversed depending on whether a pin or signal is implemented in a positive or negative circuit system, etc. In some cases, the circuit system state may be programmable, wherein the circuit system state may be reversed for a given circuit system function.
[0078] As used herein, the term "a" is defined as one or more than one. Furthermore, the use of introductory phrases such as "at least one" and "one or more" in the claims should not be interpreted as implying that the introduction of another claim element by the indefinite article "a" will limit any particular claim containing this introduced claim element to an invention containing only one such element, even when the same claim includes the introductory phrases "one or more" or "at least one" and indefinite articles such as "a". The same applies to the use of definite articles. Unless otherwise stated, terms such as "first" and "second" are used to arbitrarily distinguish the elements described by such terms. Therefore, these terms do not necessarily wish to indicate the temporal priority or other priority ordering of these elements.
Claims
1. An analog / digital converter, characterized in that: include: a digital-to-analog converter (DAC) including a sampling node coupled to a capacitor array that receives a digital control input; a comparator coupled to compare the voltage of the sampling node with a reference voltage to generate an input comparison value; as well as A controller is coupled to preset the digital control input with a preset value, prompt the DAC to sample an input voltage onto the sampling node, reset the digital control input, and then perform conversions to determine a digital output by continuously adjusting the digital control input based on the input compare value.
2. The analog-to-digital converter according to claim 1, characterized in that Also includes: a sample predictor coupled to provide a predicted sample value based on the input voltage; and Wherein the controller is configured to select the preset value based on the predicted sample value.
3. The analog-to-digital converter according to claim 1, characterized in that: The capacitor array includes a plurality of capacitors, each capacitor having a first terminal coupled to the sampling node and each capacitor having a second terminal coupled to an output of a corresponding one of a plurality of drivers, wherein each of the plurality of drivers has an input that receives a corresponding one of a plurality of bits of the digital control input, and wherein each of the plurality of drivers is coupled to drive a corresponding one of the plurality of capacitors to a selected one of first and second differential reference voltages based on a corresponding one of the plurality of bits.
4. The analog-to-digital converter according to claim 1, wherein: The DAC has a full-scale range between an upper reference voltage level and a lower reference voltage level, and wherein the controller is configured to select the preset value so that the voltage of the sampling node is adjusted to a voltage between the upper reference voltage level and the lower reference voltage level when the digital control input is reset.
5. The analog-to-digital converter according to claim 1, characterized in that: The controller includes a range extension controller coupled to preset the digital control input with a preset value, prompt the DAC to sample the input voltage onto the sampling node, reset the digital control input, and then prompt the conversion; and a successive approximation register controller coupled to perform the conversion by continuously adjusting the digital control input based on the input compare value to determine the digital output when a prompt is received.
6. A method for converting an analog input voltage into a digital value, characterized in that include: Presetting a digital control input of a digital / analog converter (DAC) with a preset value; Sampling the input voltage to a sampling node of the DAC; resetting the digital control input of the DAC; comparing the voltage of the sampling node with a reference voltage to generate an input comparison value; as well as Conversion is performed by continuously adjusting the digital control input based on the input comparison value to determine a digital output.
7. The method according to claim 6, characterized in that Also included is subtracting a digital adjustment value indicative of the preset value from the digital output to provide an adjusted digital output.
8. The method according to claim 6, characterized in that Also includes: Predict the next sample value; and The preset value is determined based on the predicted sample value.
9. The method according to claim 8, characterized in that The determining the preset value includes determining the preset value so that the voltage of the sampling node is adjusted to a voltage within a full-scale range of the DAC when the digital control input of the DAC is reset.
10. The method according to claim 6, characterized in that Also includes: configuring the DAC as a plurality of capacitors, each capacitor having a first terminal coupled to the sampling node and each capacitor having a second terminal coupled to an output of a respective one of a plurality of drivers; providing each of the plurality of drivers with an input receiving a corresponding one of the plurality of bits of the digital control input; and configuring each of the plurality of drivers to drive a corresponding one of the plurality of capacitors to a selected one of the first and second differential reference voltages based on a corresponding one of the plurality of bits.
Citation Information
Cited By
High-resolution ADC (Analog to Digital Converter) range extension method, system and device based on program control compensation
CN122151659A
