High-linearity low-power-consumption charge redistribution DAC (Digital-to-Analog Converter) capacitor array

By designing a single-sided capacitor cycle split capacitor circuit without common mode voltage VCM and low-position analog switch timing in an analog-to-digital converter, the problem of high power consumption of DAC capacitor array is solved, and the charge heavy load of high linearity and low power consumption is achieved Distribution, reduces energy loss and is suitable for low voltage applications.

CN119010897APending Publication Date: 2024-11-2258TH RES INST OF CETC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411192766.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-28
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

In analog-to-digital converters, the high power consumption problem of DAC capacitor arrays, especially the energy loss caused by the common mode voltage VCM, affects the performance and efficiency of the ADC.

Method used

A single-sided capacitor cyclic split capacitor circuit without the common mode voltage VCM is designed, and the common mode voltage VCM is generated by charge transfer, and the single-sided capacitor charge and discharge is controlled through the low-position analog switching timing to replace the charging and discharge of the double-sided capacitor and reduce energy. loss.

Benefits of technology

High linearity and low power consumption charge redistribution is achieved, reducing energy loss by 72.1% compared to the traditional method and avoiding additional static power consumption, suitable for low voltage applications.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119010897A_ABST
    Figure CN119010897A_ABST
Patent Text Reader

Abstract

The invention discloses a high-linearity low-power-consumption charge redistribution DAC capacitor array, and belongs to the field of digital-to-analog converters. According to the single-side capacitor circulation split capacitor circuit without common-mode voltage VCM, the common-mode voltage VCM is generated through charge transfer based on existing reference voltage Vref, and charging and discharging of a single-side capacitor are controlled through a low-level analog switch time sequence instead of simultaneous charging and discharging of a double-side capacitor. Compared with the traditional low-position analog switch sequence, 72.1% of energy loss can be reduced; and on the other hand, the switch design of a time domain integral structure and the like needing VCM voltage generally has lower switch power consumption, but the generation of the VCM needs a band-gap reference circuit, a voltage buffer and the like, so that higher static power consumption is directly caused. The capacitor array is divided into two sub-capacitor arrays to generate the VCM, so that static power consumption generated by an extra band-gap reference circuit is avoided.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of digital-to-analog converters, and in particular to a DAC capacitor array with high linearity, low power consumption and charge redistribution. Background Art

[0002] Digital-to-analog converters are widely used in mixed signal systems such as modern communication systems, audio and video processing systems, etc. Their main function is to convert the digital signal input into the corresponding analog signal, and the DAC structure is also widely used in the internal structure of the analog-to-digital converter.

[0003] In analog-to-digital converters, the accuracy and linearity of the DAC capacitor array directly affect the performance of the entire ADC, and its importance is self-evident. In terms of structural selection, the capacitor array DAC determines the sampling rate of the SAR ADC. The capacitance value of the capacitor array DAC should be reduced as much as possible to maximize energy efficiency. Therefore, a segmented capacitor array with the least number of unit capacitors is selected. For switching timings such as VCM-based, Tri-level, and Sub-merging, although zero power consumption can be achieved when judging the highest bit, a new common-mode voltage VCM and analog drive are introduced, resulting in more energy loss. Therefore, it is necessary to design a DAC capacitor array structure to solve the above problems. Summary of the invention

[0004] The object of the present invention is to provide a DAC capacitor array with high linearity and low power consumption and charge redistribution to solve the problems in the background technology.

[0005] In order to solve the above technical problems, the present invention provides a DAC capacitor array with high linearity and low power consumption and charge redistribution, which does not require a common mode voltage V CM The unilateral capacitor cycle split capacitor circuit is based on the existing reference voltage V ref , using charge transfer to generate a common-mode voltage V CM , the charging and discharging of single-sided capacitors is controlled by low-bit analog switch timing instead of charging and discharging of bilateral capacitors at the same time.

[0006] In one embodiment, the common mode voltage V CM The unilateral capacitor cycle split capacitor circuit includes a positive terminal capacitor array, a negative terminal capacitor array, and a comparator;

[0007] The positive end capacitor array and the negative end capacitor array each include four capacitors C0, and the upper plates of two capacitors in the positive end capacitor array are connected to V ref , denoted as C P_1 The upper plates of the two capacitors in the other half of the positive capacitor array are grounded, denoted as C P_2 ; The upper plates of the two capacitors in the negative terminal capacitor array are connected to V ref, denoted as C N_1 The upper plates of the two capacitors in the other half of the negative terminal capacitor array are grounded, denoted as C N_2 ; All the capacitor upper plates of the positive terminal capacitor array are connected to the positive terminal of the comparator, and all the capacitor upper plates of the negative terminal capacitor array are connected to the negative terminal of the comparator; if V in >0, the lower plate of the negative capacitor array is grounded during the entire conversion process, and the negative input V XN During the whole process, there will be no change, and the negative terminal capacitor array will not generate energy loss; if V in <0, during the entire conversion phase, the positive terminal capacitor array will not generate energy loss, and the positive terminal input V XP Always remain the same.

[0008] In one embodiment, the switching timing of the low-bit analog is based on the high-linearity lower plate sampling to propose a split capacitor low-bit analog comparison idea, wherein the capacitor array split structure is [8C0, 8C0,, 4C0, 4C0, 2C0, 2C0, 1C0, 1C0, 1C0, 1C0], and the above split capacitor array has two sub-capacitor arrays to form a positive capacitor and a negative capacitor array: C P_1 and C P_2 Forming a positive terminal capacitor array, C N_1 and C N_2 The capacitors in the sub-capacitor array are used to replace the large capacitors for charging and discharging, which can achieve energy reuse and reduce the energy loss of large capacitor charging and discharging; the capacitors C in the positive capacitor array P2_1 With C P2_2 and the negative terminal capacitor C N2_1 With C N2_2 The small SARADC is selected to perform a binary search on the input signal, simulating the conversion operation of a large capacitor.

[0009] In one embodiment, during the conversion operation of simulating a large capacitor, the first conversion determines that the MSB still does not consume energy, the second conversion uses a unit capacitor to replace the MSB capacitor for bit cycling, and the third conversion directly switches the main capacitor array based on the conversion comparison results of the first two small SARADCs, further reducing the energy loss of the high-order capacitor during the conversion comparison process.

[0010] The present invention provides a high-linearity, low-power charge redistribution DAC capacitor array. The low-bit analog switch sequence can reduce 72.1% of energy loss compared to the traditional one. Although the monotonic switch switching has lower power consumption than the present invention, its common mode is continuously reduced and is not suitable for low-voltage applications. On the other hand, the time domain integration structure requires V CMVoltage switching designs generally have lower switching power consumption, but V CM The generation of V requires a bandgap reference circuit and a voltage buffer, which directly leads to greater static power consumption. The present invention splits the capacitor array into two sub-capacitor arrays to generate V CM , thereby avoiding the static power consumption generated by the additional bandgap reference circuit. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a schematic diagram of generating common-mode voltage VCM by charge redistribution.

[0012] Figure 2 This is a schematic diagram of a split capacitor design that does not require VCM.

[0013] Figure 3 It is a schematic diagram of a single-sided capacitor cycle split capacitor design that does not require VCM proposed by the present invention.

[0014] Figure 4 It is a schematic diagram of the MSB-2 determination method proposed by the present invention.

[0015] Figure 5 It is a schematic diagram of a split capacitor array using low-bit simulation proposed by the present invention.

[0016] Figure 6 It is a schematic diagram of the power consumption analysis of the first three switches of the low-bit simulation judgment. DETAILED DESCRIPTION

[0017] The following is a further detailed description of a high linearity, low power consumption charge redistribution DAC capacitor array proposed by the present invention in conjunction with the accompanying drawings and specific embodiments. The advantages and features of the present invention will become clearer according to the following description. It should be noted that the accompanying drawings are all in a very simplified form and are not in precise proportions, and are only used to conveniently and clearly assist in explaining the purpose of the embodiments of the present invention.

[0018] The present invention first utilizes the existing reference voltage V ref Based on the essence of charge conservation, charge transfer is used to generate a common-mode voltage V CM The basic idea is as follows Figure 1 As shown, connect the two ends of a capacitor C0 to V ref and ground, and both ends of another capacitor C0 of the same size are grounded. The upper plates of the two capacitors are connected and the lower plates are grounded. According to the law of charge conservation, after charge redistribution:

[0019]

[0020] This is equivalent to connecting one end of the capacitor to V CM , the other end is grounded. Based on this idea, the introduction of a new common-mode voltage V is avoided. CMAnd related analog drivers. Based on the traditional 3-bit SARADC capacitor array, V CM -based switch timing, increase the highest capacitor 2C0 and split it into two unit capacitors, and then follow Figure 1 The idea shown is that the upper plate of half of the capacitors of the positive and negative capacitor arrays is connected to V ref , the other half of the capacitor upper plate is grounded, and then all the positive end capacitor upper plates are connected to the positive terminal of the comparator, and all the negative end capacitor upper plates are connected to the negative terminal of the comparator. The common mode level is generated by charge redistribution, such as Figure 2 As shown, after the process is completed, according to the charge conservation principle, the comparator positive and negative input V XP 、V XN They are:

[0021] V XP =1 / 2V ref -V inp #(2)

[0022] V XN =1 / 2V ref -V inn #(3)

[0023] Formulas (2) and (3) show that: the first bit cycle input voltage V in (V in =V XP -V XN ) is compared with 0 to determine the highest bit. There is no charge transfer in this process, so no energy loss is generated when determining the highest bit.

[0024] exist Figure 3 In order to further reduce energy loss, single-side capacitor charging and discharging is used instead of double-side capacitor charging and discharging simultaneously. in >0, the lower plate of the negative capacitor array is grounded during the entire conversion process, that is, the negative input V XN will not change during the whole process, so the negative capacitor array will not generate energy loss; similarly, if V in <0, the positive capacitor array will not generate energy loss during the entire conversion phase, and the positive input V XP Always remain unchanged. Here we assume that V in >0, the negative capacitor array is grounded during the entire conversion process, and the negative input of the comparator is V XN Keep V XN =1 / 2V ref -V inn For simplicity, Figure 4 The negative terminal capacitor array structure is no longer drawn.

[0025] Undoubtedly, compared with the simultaneous circulation of two-terminal capacitors, the use of single-sided capacitor array circulation can reduce power consumption, but the problem that still needs to be considered is how to generate the voltage level required by the binary search algorithm, such as 1 / 2V, by the single-sided capacitor circulation. ref , 1 / 4V ref , 3 / 4V ref Etc. The following example illustrates that assuming V in >1 / 2V ref , How does a capacitor array generate 3 / 4V ref situation so that V in With 3 / 4V ref Make a comparison.

[0026] Single-sided capacitor array generates 3 / 4V ref have Figure 4 The two paths are shown. Figure 4 The effect of the switching sequence on power consumption is revealed: the first path at the top connects the lower analog capacitor to the rest of the main capacitor array, and then connects the lower plates of the two capacitors to V ref The second path below is to connect the lower plates of the remaining capacitors in the main capacitor array to V ref , and then connect the two capacitor arrays. Through observation and estimation, the power consumption of the upper path 1 is the charge flowing out of the power supply and V ref The product is obtained by multiplying the charges released from the capacitor and flowing into the power supply, while the charges released from the capacitor and flowing into the power supply are wasted. In the lower path 2, the charges released from the capacitor and flowing into the power supply are partially offset by the charges flowing out of the power supply, thereby reducing power consumption. Next, theoretical calculations and verification are performed.

[0027] like Figure 4 As shown, the power consumption of each switching step is as follows:

[0028] ①Power consumption:

[0029]

[0030] ②Power consumption:

[0031]

[0032] ③Power consumption:

[0033]

[0034] ④Power consumption:

[0035]

[0036] ⑤Power consumption:

[0037]

[0038] The calculations (4) to (8) above show that the power consumption of the first path (①②③) is The power consumption of the second path (④⑤) is That is, the power consumption of switching paths ①②③ is higher than that of switching paths ④⑤, which is consistent with the previous observations. Therefore, compared with the first path, the second path can reduce the power consumption by optimizing the switching timing. The power consumption shows the importance of switch switching design.

[0039] for Figure 4 According to the comparison result of the previous comparator, only the positive terminal capacitor array C P_1 or C P_2 The capacitors in the conventional capacitor array are charged and discharged during conversion and comparison, while all capacitors are completely charged and discharged, resulting in energy waste. The split capacitor unilateral circulation design proposed in the present invention realizes more efficient energy storage and reuse in the conversion stage. Taking a 6-bit SAR ADC as an example, the quantitative analysis theory of power consumption is derived as follows:

[0040] ① Determine the power consumption of the highest bit MSB: 0

[0041] ② Determine the power consumption of the second highest bit MSB-1:

[0042] ③ Determine the power consumption of MSB-2:

[0043]

[0044] By analogy, the power consumption of each bit is reduced to 1 / 4 of the previous bit, so the total power consumption of 6-bit SAR ADC conversion is: That is, compared with the traditional switch switching method, 66.2% of power consumption is saved.

[0045] In the above capacitor splitting structure, the energy loss is successfully reduced by relying only on the single-sided capacitor array bit cycle. If further reducing the power consumption is considered, two unit capacitors can be used in the positive and negative capacitor arrays of the capacitor splitting structure to form a two-bit SAR ADC to simulate the high-bit comparison process of the overall SAR ADC.

[0046] like Figure 5 As shown, taking a 6-bit SAR ADC as an example, the positive and negative capacitor arrays of the traditional 6-bit SAR ADC are [32C0, 16C0, 8C0, 4C0, 2C0, C0, C0] split into sub-capacitor arrays, and according to the previous analysis, the improved capacitor array split structure of the present invention is [8C0, 8C0, 4C0, 4C0, 2C0, 2C0, 1C0, 1C0, 1C0, 1C0], and the above split capacitor arrays respectively have two sub-capacitor arrays to form the positive capacitor and the negative capacitor, that is, C P_1 and CP_2 Forming a positive terminal capacitor array, C N_1 and C N_2 The capacitors in the sub-capacitor array are used to replace the large capacitors for charging and discharging, which can achieve energy reuse and reduce the energy loss of large capacitor charging and discharging.

[0047] Figure 5 The capacitor C in the positive terminal capacitor array P2-1 With C P2-2 and the negative terminal capacitor C N2-1 With C N2-2 The selected small SARADC performs binary search on the input signal to simulate the conversion operation of the large capacitor. In the process of using the sub-SARADC to perform bit cycle (i.e., perform binary search to simulate the conversion operation of the large capacitor), the first conversion to judge the MSB still does not consume energy, the second conversion uses the unit capacitor to replace the MSB capacitor for bit cycle, and the third conversion will directly switch the main capacitor array according to the conversion comparison results of the first two small SARADCs, thereby further reducing the energy loss of the high-bit capacitor during the conversion comparison process.

[0048] Figure 6 The power consumption generated by the low-bit analog capacitor switch switching method proposed in the present invention is summarized. The first step is sampling. After the sampling is completed, the connection with Vref and ground is disconnected, and the lower plate of the capacitor is grounded. No charge flows out of the power supply during the grounding process, so E DAC =0. According to the result of the first step comparator, the reference voltage connection mode of the sub-capacitor array is set. If the comparator judgment result is 1, the comparator positive terminal capacitor array is connected to Vref, otherwise the comparator negative terminal capacitor array is connected to Vref. The energy consumed in this step is: Afterwards, the sub-capacitor array is connected back to the main capacitor array according to the sub-capacitor array voltage comparison result and the connection mode of Vref is adjusted to complete the next comparison. This step consumes In summary, the average power consumption required to judge the first three is The traditional binary capacitor switch switching method determines the power consumption of the first three digits as

[0049] The above description is only a description of the preferred embodiments of the present invention, and is not intended to limit the scope of the present invention. Any changes or modifications made by a person skilled in the art in the field of the present invention based on the above disclosure shall fall within the scope of protection of the claims.

Claims

1. A high linearity, low power consumption charge redistribution DAC capacitor array, characterized in that: No common mode voltage V is required CM The unilateral capacitor cycle split capacitor circuit is based on the existing reference voltage V ref , using charge transfer to generate a common-mode voltage V CM , the charging and discharging of single-sided capacitors is controlled by low-bit analog switch timing instead of charging and discharging of bilateral capacitors at the same time.

2. The high linearity, low power consumption, charge redistribution DAC capacitor array according to claim 1, characterized in that: The common mode voltage V CM The unilateral capacitor cycle split capacitor circuit includes a positive terminal capacitor array, a negative terminal capacitor array, and a comparator; The positive end capacitor array and the negative end capacitor array each include four capacitors C0, and the upper plates of two capacitors in the positive end capacitor array are connected to V ref , denoted as C P_1 The upper plates of the two capacitors in the other half of the positive terminal capacitor array are grounded, denoted as C P_2 ; The upper plates of the two capacitors in the negative terminal capacitor array are connected to V ref , denoted as C N_1 , the upper plates of the two capacitors in the other half of the negative terminal capacitor array are grounded, denoted as C N_2 ; All the capacitor upper plates of the positive terminal capacitor array are connected to the positive terminal of the comparator, and all the capacitor upper plates of the negative terminal capacitor array are connected to the negative terminal of the comparator; if V in >0, the lower plate of the negative capacitor array is grounded during the entire conversion process, and the negative input of the comparator V XN During the whole process, there will be no change, and the negative terminal capacitor array will not generate energy loss; if V in <0, during the entire conversion phase, the positive terminal capacitor array will not generate energy loss, and the positive terminal input V XP Always remain the same.

3. The high linearity, low power consumption charge redistribution DAC capacitor array according to claim 2, characterized in that: The switching timing of the low-bit simulation is based on the high-linearity lower plate sampling and proposes the idea of ​​splitting the capacitor low-bit simulation comparison, wherein the capacitor array split structure is [8C0, 8C0,, 4C0, 4C0, 2C0, 2C0, 1C0, 1C0, 1C0, 1C0], and the above split capacitor array has two sub-capacitor arrays to form the positive end capacitor and the negative end capacitor array: C P_1 and C P_2 Forming a positive terminal capacitor array, C N_1 and C N_2 The capacitors in the sub-capacitor array are used to replace the large capacitors for charging and discharging, which can achieve energy reuse and reduce the energy loss of large capacitor charging and discharging; the capacitor C in the positive capacitor array P2_1 With C P2_2 and the negative terminal capacitor C N2_1 With C N2_2 The small SARADC is selected to perform a binary search on the input signal, simulating the conversion operation of a large capacitor.

4. The high linearity, low power consumption charge redistribution DAC capacitor array according to claim 3, characterized in that: In the process of simulating the conversion operation of large capacitors, the first conversion determines that the MSB still does not consume energy. The second conversion uses the unit capacitor to replace the MSB capacitor for bit cycling. The third conversion directly switches the main capacitor array based on the conversion comparison results of the first two small SARADCs, further reducing the energy loss of high-bit capacitors during the conversion and comparison process.