A segmented capacitor array adjustment method for increasing accuracy in SAR ADCs by utilizing partially redundant capacitors.

Abstract

The application belongs to the technical field of electronics, and discloses a segmented capacitor array adjusting method for increasing precision by utilizing partial redundant capacitors in a SAR ADC, comprising a segmented capacitor array, the method effectively utilizes the redundant amount by adjusting the sizes of the bridging capacitors and Cd capacitors of the segmented capacitor array, and increases the ENOB of the ADC. By reasonably adjusting the sizes of the bridging capacitors and Cd capacitors of the segmented capacitor array, the redundant amount is effectively utilized, more power consumption and area are not increased, the conversion times are not increased, and the ENOB of the ADC is increased.

Description

Technical Field

[0001] This invention belongs to the field of electronic technology, and in particular relates to a segmented capacitor array adjustment method for use in SAR ADCs, which can increase accuracy by utilizing partially redundant capacitors. Background Technology

[0002] Capacitor arrays are primarily used in ADCs as CDAC arrays, where they work in conjunction with comparators to convert input analog signals into digital signals. Redundant segmented capacitor arrays divide the capacitor array into multiple segments, each connected by bridging capacitors. A redundant bit with the same weight as the highest bit in each segment (except the highest-weighted segment) is added. Compared to ordinary binary capacitor arrays, redundant segmented capacitor arrays can reduce the capacitor array area and mitigate the impact of errors caused by non-ideal factors.

[0003] Taking a three-segment capacitor array as an example, each segment, from high to low, is the HSB (High Significant Bit) segment, MSB (Middle Significant Bit) segment, and LSB (Low Significant Bit) segment. The setting of the bridging capacitor is crucial, as its value affects the weight of each bit in the capacitor array. The usual rule is to ensure that during the conversion stage, the charge distribution between the two capacitors at the ends of the bridging capacitor is in a 2:1 ratio. However, since the unit capacitance is fixed, the calculated number of bridging capacitors is often in fractional form. This is where the Cd capacitor, connected between the MSB (or LSB) segment and ground, comes in, and can be used to distribute the charge within the segment. By setting the bridging capacitor to an integer multiple of the unit capacitance and then adjusting the size of the Cd capacitor to an appropriate value, the weight values ​​of the capacitor array can be made binary.

[0004] Capacitor arrays set up in this way typically suffer from resolution issues due to capacitor mismatch. Even after calibration, their resolution will be lower than ideal, and the added redundant capacitors are generally not fully utilized, resulting in some waste. Summary of the Invention

[0005] The purpose of this invention is to provide a segmented capacitor array adjustment method for use in SAR ADCs, which can increase accuracy by utilizing partially redundant capacitors, in order to solve the above-mentioned technical problems.

[0006] To address the aforementioned technical problems, the present invention provides a specific technical solution for a segmented capacitor array adjustment method applied to SAR ADCs, which utilizes partially redundant capacitors to increase accuracy:

[0007] A segmented capacitor array adjustment method for increasing accuracy in SAR ADCs by utilizing partially redundant capacitors includes a segmented capacitor array. The method effectively utilizes the excess redundancy by adjusting the size of the bridging capacitor and Cd capacitor of the segmented capacitor array, thereby increasing the ENOB of the ADC.

[0008] Furthermore, the segmented capacitor array is a dual-ended binary three-segment capacitor array with high-five-bit sampling and low-five-bit calibration. Each segment of the capacitor array, from high to low, is the HSB segment, MSB segment, and LSB segment. The HSB segment includes a capacitor C. b9 C b10 C b11 C b12r C b12 C b13 C b14 C b15 The MSB segment includes C d2 C b5 C b6 C b7 C b8r C b8 The LSB segment includes C d1 C b0 C b1 C b2 C b3 C b4r C b4 , where C b4r C b8r C b12r As redundant capacitors, C is changed by adjusting the bridging capacitor between each segment. b5 and C b4 The weight ratio, C b9 and C b8 The weighting ratio is adjusted to reduce the voltage value corresponding to 1 LSB, thereby increasing the resolution.

[0009] Furthermore, the changes in the capacitor array after adjusting the bridging capacitor were analyzed. Under ideal conditions, for segments Cb5 and LSB, when the bridging capacitor is smaller than the ideal bridging capacitor size, as long as the sum of the weights of the LSB segments plus the weight corresponding to one LSB is not lower than the weight of Cb5, it means that successive approximation can still be performed correctly, and the result can be calibrated using a bottom-up calibration method. If the sum of the weights of the LSB segments plus the weight corresponding to one LSB is lower than the weight of Cb5, that is, the redundancy has been used up, then a mode loss error will occur. When the bridging capacitor is larger than the ideal bridging capacitor size, the voltage corresponding to the LSB increases, the resolution decreases, and a code loss error occurs.

[0010] Furthermore, by reducing the voltage value corresponding to 1 LSB without changing the full-scale voltage, and with the code value range being 0 / LSB to (2Vref-LSB) / LSB, i.e., reducing the voltage value corresponding to 1 LSB, the LSB resolution can be increased compared to the ideal bridge capacitor case.

[0011] Furthermore, the changes in the set bridge capacitor size and weight after altering the bridge capacitor are calculated, adjusting both bridge capacitors to just utilize the redundancy:

[0012] at this time,

[0013] sum(W real(9:15) ) = sum(W ideal(9:15) )

[0014] sum(W real(5:8) ) = W real(9) -W real(5) (5:8 includes b8r)

[0015] sum(W real(0:4) ) = W real(5) -W real(0) (0:4 includes b4r)

[0016]

[0017] have to,

[0018]

[0019] Right now,

[0020] C br1 =C b0

[0021] because,

[0022]

[0023] Right now,

[0024]

[0025] And because,

[0026] W real(9) =W ideal(9) =512

[0027] Right now,

[0028]

[0029] That is, the weight corresponding to one LSB is as shown in the formula above. Compared with the voltage value corresponding to one LSB before adjusting the bridge capacitor, it is reduced by a factor of 2.25.

[0030] In ideal conditions, ENOB could use all its redundancy to increase resolution to 17.169925 bits. Furthermore, adding C... d Capacitor, adjust the bridging capacitor to an integer multiple of the unit capacitor, and adjust the bridging capacitor and C. d The capacitor size ratio is adjusted so that, while ensuring that both capacitors are integer multiples of the unit capacitance, the excess redundancy is used to increase resolution.

[0031] Furthermore,

[0032] Introducing C under ideal conditions d Capacitor, assuming C br1 =C br2 =2C b0 ENOB = 17.169925.

[0033] at this time,

[0034]

[0035] have to,

[0036] C d1 =47C b0

[0037] Similarly

[0038]

[0039] We can obtain,

[0040]

[0041] C d1 It is also an integer multiple of the unit capacitance, but C d2 To adjust C to a fractional multiple of the unit capacitance, continue adjusting C. br2 Or C br1 C d2 The values ​​are adjusted to integer multiples of the unit capacitance.

[0042] The segmented capacitor array adjustment method of the present invention, which utilizes partially redundant capacitors to increase accuracy in SAR ADCs, has the following advantages: The present invention effectively utilizes the excess redundancy by reasonably adjusting the size of the bridging capacitor and Cd capacitor of the segmented capacitor array, without increasing power consumption and area, or increasing the number of conversions, thereby increasing the ENOB of the ADC. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the binary three-segment capacitor array structure with high-five-bit sampling and low-five-bit calibration according to the present invention.

[0044] Figure 2 This is a schematic diagram of the V-code curve structure under different bridging capacitor sizes according to the present invention. Detailed Implementation

[0045] To better understand the purpose, structure, and function of this invention, the following description, in conjunction with the accompanying drawings, provides a more detailed account of a segmented capacitor array adjustment method for increasing accuracy in SAR ADCs using partially redundant capacitors.

[0046] This invention increases the ENOB of the ADC by reasonably adjusting the size of the bridging capacitor and Cd capacitor in the segmented capacitor array, effectively utilizing the excess redundancy, without increasing power consumption or area, or increasing the number of conversions.

[0047] The technical solution proposed in this invention is mainly applied to segmented capacitor arrays with redundancy. As shown in Figure 1, taking one segment of a dual-ended binary three-segment capacitor array with high-five-bit sampling and low-five-bit calibration as an example, each segment of the capacitor array, from high to low, is the HSB segment, MSB segment, and LSB segment. The HSB segment includes capacitor C. b9 C b10 C b11 C b12r C b12 C b13 C b14 C b15 The MSB segment includes C d2 C b5 C b6 C b7 C b8r C b8 The LSB segment includes C d1 C b0 C b1 C b2 C b3 C b4r C b4 , where C b4r C b8r C b12r Assuming the full-scale range of Vip-Vin is -Vref to +Vref, the corresponding code range is 0 to 65535. Due to the redundancy, the actual code range can reach -2184 to 67719. However, since the full-scale voltage cannot exceed the range, the output code value cannot reach outside the range under ideal conditions, that is, it can only be within 0 to 65535.

[0048] In fact, C can be changed by adjusting the bridging capacitor between each segment. b5and C b4 weight ratio of, C b9 and C b8 weight ratio of, thereby reducing the voltage value corresponding to 1 LSB and increasing the resolution.

[0049] The following analyzes the changes in the capacitor array after adjusting the bridging capacitor. Taking the ideal situation, Cb5 and the LSB segment as an example, when the bridging capacitor is smaller than the ideal bridging capacitor size, as long as the sum of the weights of the LSB segment plus the weight corresponding to 1 LSB is not lower than the weight of Cb5, it means that the successive approximation can still be carried out correctly, and the obtained result can be calibrated through the bottom-up calibration method; if the sum of the weights of the LSB segment plus the weight corresponding to 1 LSB is lower than the weight of Cb5, that is, the redundancy has been used up, then a modeling error will occur at this time. When the bridging capacitor is larger than the ideal bridging capacitor size, the voltage corresponding to the LSB increases, the resolution decreases, and a coding error appears at the same time.

[0050] There are a total of four cases here, corresponding respectively to Figure 2 the V-code curves under different bridging capacitor sizes. Among them, C_bridge < C_ideal (just using up the redundancy) is the result of the design direction of this invention under ideal conditions, and the redundancy can be used to increase the resolution.

[0051] By reducing the voltage value corresponding to 1 LSB, and the full-scale voltage remains unchanged, and the code value range is 0 / LSB to (2Vref - LSB) / LSB, that is, by reducing the voltage value corresponding to 1 LSB, the LSB resolution can be increased compared with the ideal bridging capacitor situation.

[0052] Next, calculate the change in the set bridging capacitor size and weight after changing the bridging capacitor.

[0053] Adjust both bridging capacitors to the situation where the redundancy is just used up:

[0054] At this time,

[0055] sum(W real(9:15) ) = sum(W ideal(9:15) )

[0056] sum(W real(5:8) ) = W real(9) - W real(5) (5:8 includes b8r)

[0057] sum(W real(0:4) ) = W real(5) - W real(0) (0:4 includes b4r)

[0058]

[0059] We can obtain,

[0060]

[0061] Right now,

[0062] C br1 =C b0

[0063] because,

[0064]

[0065] Right now,

[0066]

[0067] And because,

[0068] W real(9) =W ideal(9) =512

[0069] Right now,

[0070]

[0071] That is, the weight of one LSB is as shown in the above formula, which is 2.25 times smaller than the voltage value of one LSB before adjusting the bridge capacitor.

[0072] In other words, under ideal conditions, ENOB can use all its redundancy to increase the resolution to 17.169925 bits.

[0073] This is under the extremely ideal condition, with the bridging capacitor C... br1 It is exactly equal to the unit capacitance, but C br2 Using fractional multiples of the unit capacitance is disadvantageous for setting the number of components and for matching. A Cd capacitor can be added to adjust the bridge capacitance to an integer multiple of the unit capacitance. This is because adding a Cd capacitor serves the same purpose as adjusting the bridge capacitance: changing the weight of the transition bits or altering their charge distribution. This reduces the voltage corresponding to the LSB or MSB, increasing resolution. Therefore, by appropriately adjusting the ratio of the bridge capacitor to the Cd capacitor, while ensuring both capacitors are integer multiples of the unit capacitance, it is also possible to utilize redundancy to increase resolution.

[0074] Taking an ideal case as an example, let's introduce a Cd capacitor, assuming... C br1 =C br2 =2C b0 ENOB = 17.169925.

[0075] at this time,

[0076]

[0077] We can obtain,

[0078] C d1 =47C b0

[0079] Similarly

[0080]

[0081] We can obtain,

[0082]

[0083] As can be seen, C d1 It is also an integer multiple of the unit capacitance, but C d2 This is a fractional multiple of the unit capacitance; C can be further adjusted here. br2 Or C br1 C d2 The values ​​of these are adjusted to integer multiples of the unit capacitance; the derivation will not be detailed here.

[0084] It is understood that the present invention has been described through some embodiments, and those skilled in the art will recognize that various changes or equivalent substitutions can be made to these features and embodiments without departing from the spirit and scope of the invention. Furthermore, under the teachings of the present invention, these features and embodiments can be modified to adapt to specific situations and materials without departing from the spirit and scope of the invention. Therefore, the present invention is not limited to the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the protection scope of the present invention.

Claims

1. A segmented capacitor array adjustment method for increasing accuracy in SAR ADCs by utilizing partially redundant capacitors, comprising a segmented capacitor array, characterized in that, The method effectively utilizes excess redundancy and increases the ENOB of the ADC by adjusting the size of the bridging capacitor and Cd capacitor in the segmented capacitor array. Calculate the changes in the set bridge capacitor size and weight after changing the bridge capacitor. Adjust both bridging capacitors to a point where the redundancy is just fully utilized: at this time, sum(W real(9:15) )=sum(W ideal(9:15) ) sum(W real(5:8) ) = W real(9) -W real(5) (5:8 includes b8r) sum(W real(0:4) ) = W real(5) -W real(0) (0:4 includes b4r) have to, Right now, C br1 =C b0 because, Right now, And because, IN real(9) =In ideal(9) =512 Right now, The weight of one LSB is as shown in the formula above. Compared with the voltage value of one LSB before adjusting the bridge capacitor, it is reduced by 2.25 times. That is, under ideal conditions, ENOB can use all the redundancy to increase the resolution to 17.169925 bits.

2. The segmented capacitor array adjustment method for increasing accuracy in SAR ADCs using partially redundant capacitors, as described in claim 1, is characterized in that... The segmented capacitor array is a dual-ended binary three-segment capacitor array with high-five-bit sampling and low-five-bit calibration. Each segment of the capacitor array, from high to low, is the HSB segment, MSB segment, and LSB segment. The HSB segment includes a capacitor C. b9 C b10 C b11 C b12r C b12 C b13 C b14 C b15 The MSB segment includes C d2 C b5 C b6 C b7 C b8r C b8 The LSB segment includes C d1 C b0 C b1 C b2 C b3 C b4r C b4 , where C b4r C b8r C b12r As redundant capacitors, C is changed by adjusting the bridging capacitor between each segment. b5 and C b4 The weight ratio, C b9 and C b8 The weighting ratio is adjusted to reduce the voltage value corresponding to 1 LSB, thereby increasing the resolution.

3. The segmented capacitor array adjustment method for increasing accuracy in SAR ADCs using partially redundant capacitors, as described in claim 2, is characterized in that... Analyzing the changes in the capacitor array after adjusting the bridging capacitor reveals that, ideally, for segments Cb5 and LSB, when the bridging capacitor is smaller than the ideal bridging capacitor size, as long as the sum of the weights of the LSB segments plus the weight corresponding to one LSB is not lower than the weight of Cb5, the successive approximation can still proceed correctly, and the result can be calibrated using a bottom-up calibration method. If the sum of the weights of the LSB segments plus the weight corresponding to one LSB is lower than the weight of Cb5, meaning the redundancy has been exhausted, then a mode loss error will occur. When the bridging capacitor is larger than the ideal bridging capacitor size, the voltage corresponding to the LSB increases, the resolution decreases, and a code loss error occurs.

4. The segmented capacitor array adjustment method for increasing accuracy in SAR ADCs using partially redundant capacitors, as described in claim 3, is characterized in that... By reducing the voltage value corresponding to 1 LSB without changing the full-scale voltage, and with the code value range being 0 / LSB to (2Vref-LSB) / LSB, the LSB resolution can be increased compared to the ideal bridge capacitor case.

5. The segmented capacitor array adjustment method for increasing accuracy in SAR ADCs using partially redundant capacitors, as described in claim 4, is characterized in that... Add C d Capacitor, adjust the bridging capacitor to an integer multiple of the unit capacitor, and adjust the bridging capacitor and C. d The capacitor size ratio is adjusted so that, while ensuring that both capacitors are integer multiples of the unit capacitance, the excess redundancy is used to increase resolution.

6. The segmented capacitor array adjustment method for increasing accuracy in SAR ADCs using partially redundant capacitors, as described in claim 5, is characterized in that... Introducing C under ideal conditions d Capacitor, assuming C br1 =C br2 =2C b0 ENOB = 17.169925. at this time, have to, C d1 =47C b0 Similarly We can obtain, C d1 It is also an integer multiple of the unit capacitance, but C d2 To adjust C to a fractional multiple of the unit capacitance, continue adjusting C. br2 Or C br1 C d2 The values ​​are adjusted to integer multiples of the unit capacitance.

Images