A method for converting analog signals to digital signals

The two-step analog-to-digital conversion method addresses the complexity and error issues in ADCs by eliminating external probes and using digital trimming, resulting in a simplified calibration process and reduced errors, achieving efficient and cost-effective ADC performance.

WO2024246576A1PCT designated stage expired Publication Date: 2024-12-05SILICON CRAFT TECH
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
PCT/IB2023/055399
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-26
Publication Date
2024-12-05

AI Technical Summary

Technical Problem

Current analog-to-digital converters (ADCs) require complex calibration processes and multiple trimmings over temperature ranges, introducing errors and increasing design complexity, despite existing methods for error correction.

Method used

A two-step analog-to-digital conversion method that eliminates the need for external analog probes, using a simple reference circuit and digital trimming to reduce error sources and simplify calibration, with digital calculations to correct errors.

Benefits of technology

This approach reduces ADC errors, simplifies the calibration process, and decreases design complexity, allowing for a single-point calibration that covers the operating temperature range, thereby reducing costs and complexity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure IB2023055399_05122024_PF_FP_ABST
    Figure IB2023055399_05122024_PF_FP_ABST
Patent Text Reader

Abstract

A method for converting analog signals to digital signals is a method to convert analog signals to digital signals in an analog-to-digital converter (ADC). The method comprises a two-step analog-to-digital conversion which can compensate for errors of the ADC reference. The method allows the use of a simple voltage reference circuit and does not require an external analog calibration probe. The method can achieve a high-accuracy ADC with a simple.
Need to check novelty before this filing date? Find Prior Art

Description

[0001]  A METHOD FOR CONVERTING ANALOG SIGNALS TO DIGITAL SIGNALS Technical field of the invention The present invention relates to a system for signal processing and more specifically to systems and methods for conversion of analog signals to digital signals, analog-to-digital converter (ADC) design, and digital compensation. Background of the invention Analog-to-digital converter (ADC) is a system that converts analog signals to digital signals, and is widely used in various electronic applications such as microcontrollers, scientific instruments, sensor signal processing, image or music recording etc. An ADC typically requires a precise voltage reference to function, which is often provided by a bandgap voltage reference circuit. This circuit provides a temperature independent voltage reference to the ADC with minimum variation over temperature and supply voltage. However, all integrated circuits (ICs) experience some variation over fabrication process, supply voltage, and temperature range, so the bandgap reference circuit requires calibration, trimming or methods to reduce errors in ADC system to account for IC process variation. To meet the stringent precision requirements of modern ADCs, the bandgap voltage reference usually needs multiple point calibrations or trimmings over the desired operation range. The errors of the ADC reference voltage can be corrected by various methods, such as direct trimming of the voltage reference circuit using external analog probing, or indirect calibration at the output through digital compensation. Although several error correction methods have been proposed in the past, they all have certain limitations. The U.S. Patent No.7,433,790B2 describes a technique for automatic reference voltage trimming that directly applies error correction to the bandgap reference circuit using an auto correction algorithm. The advantage is that it only needs external input and desired output, without the need for external probing. However, it still requires calibration at multiple temperature points and has an excessive number of design configurations, which adds to on- chip testing and support peripheral, making it more complex and occupying more area. The U.S. Patent No. 7,728,575B2 describes a method and apparatus for higher-order correction of a bandgap voltage. It applies error correction directly to the bandgap reference circuit, which improves accuracy and precision by correcting higher-order nonlinear errors. However, this invention also introduces more error sources and increases complexity. The U.S. Patent No. 9,323,274B2 describes a self-calibrating digital bandgap voltage and current reference. Although the digital bandgap reference simplifies the circuit, it requires a high-accuracy digital-to-analog converter (DAC) that needs calibration over the operating range, thereby adding complexity to the ADC. In addition, calibration of ADCs, especially over the operating temperature range, can be time consuming and costly.   Considering these limitations, there is a need for a more effective method to minimize the amount of error in analog-to-digital converters (ADCs). The present invention addresses this issue by providing a method for reducing ADC errors that does not require external analog probes and allowing a simple voltage reference circuit to be used. This results in fewer error sources, a simpler calibration process, and reduced design complexity. Summary of the invention The present invention described herein generally relates to a method for converting analog-to-digital signal (ADC) without requiring external analog calibration probes. The method involves 1) converting a first analog signal into a first digital code by using a quantizer and storing the first digital code in a digital module; 2) converting a second analog signal to be a second digital code by using the quantizer and storing the second digital code in the digital module; and3) calculating a digital output code by using the first digital code and the second digital code wherein the digital output code represents one of the said analog signals. The general purpose of our invention is to provide a method for converting analog signals to digital signals for both general-purpose analog-to-digital converter (ADC) and high- accuracy digital temperature sensor. The method for analog-to-digital conversion eliminates the need for external analog probes during calibration and uses a simple reference circuit, which reduces error sources, simplifies the calibration process, and lowers design complexity. The method involves a two-step analog-to-digital conversion (ADC) technique with fully digital trimming. Brief description of the drawing FIG.1 is an example schematic diagram of a conventional bandgap voltage reference circuit. FIG.2 is an example diagram of a quantizer. FIG. 3 is an example diagram of the analog-to-digital converter (ADC) according to the invention. FIG. 4 is an example flow chart of the method to convert analog signals to digital signals according to the invention. FIG.5 is an example flow chart of the calibration method of the analog to digital conversion (ADC) according to the invention. FIG.6 is an example diagram that illustrates the effective result of the reference voltage ( ^^^ீ) after trimming, according to the invention. Detailed description FIG. 1 shows a conventional bandgap voltage reference circuit 100 that typically consists of p-type metal oxide semiconductor field effect transistors (P-MOSFETs) 101, an operational amplifier (op-amp) 102, two resistors of a first resistor (R1) 103 and a second resistor (R2) 104, and three bipolar junction transistors (BJTs) including the first BJT 105, the second BJT 106 and the third BJT 107. The BJTs, which are biased with currents from P- MOSFETs 101, generate analog base emitter voltages ( ^^^ா^). The base-emitter voltage ^ ^^^ா^ is complementary to absolute temperature (CTAT), i.e., it decreases as the absolute temperature increases. The bipolar junction transistors (BJTs) can be replaced with semiconductor junction   diodes to generate the anode-cathode voltage instead. The op-amp 102 is arranged in a negative feedback configuration, which keeps the voltages at the positive and negative input terminals of the op-amp 102 approximately equal. This voltage is equal to a second analog base-emitter voltage ^ ^^^ாଶ). A voltage difference between the first analog base-emitter voltage ( ^^^ா^^ and the second base-emitter voltage ^ ^^^ாଶ)can be generated by having the first BJT 105 and the second BJT 106 with different emitter areas and / or with different emitter currents. The voltage difference is known as the differential base-emitter voltage (∆ ^^^ா^ which is proportional to absolute temperature (PTAT), i.e., it increases with the absolute temperature. By the circuit arrangement, the differential base-emitter voltage (∆ ^^^ா^ is equal to the voltage dropped across the first resistor (R1) 103 and this creates a current flowing through the first resistor (R1) 103. The current flowing through the first resistor (R1) 103 is proportional to absolute temperature (PTAT) and is known as the PTAT current. The P-MOSFETs 101 act as a current mirror copying the current flowing through the first resistor (R1) 103 and provide the PTAT current to the second resistor (R2) 104. The output voltage of the bandgap reference circuit, which is known as the bandgap voltage ( ^^^ீ) can then be expressed as follow: Since the differential base-emitter voltage ^∆ ^^^ா^ is proportional to absolute temperature (PTAT) and the analog base-emitter voltage ( ^^^ா) is complementary to absolute temperature (CTAT) wherein the analog base-emitter voltage ( ^^^ா) is derived by the third BJT 107, the ratio of the second resistor (R2) 104 and the first resistor (R1) 103 can be designed so that the complementary to absolute temperature (CTAT) characteristic of the analog base- emitter voltage ^ ^^^ா^ term cancels out with the proportional to absolute temperature (PTAT) characteristic of the differential base-emitter voltage ^∆ ^^^ா). This, ideally, results in the bandgap voltage ( ^^^ீ) that is independent of temperature. In reality, IC process variation of the BJTs and mismatch of resistors introduce variations in the absolute value of the analog base-emitter voltage ^ ^^^ா), and in the ratio of the second resistor (R2) 104, to the first resistor (R1) 103, respectively. This causes the bandgap voltage ^ ^^^ீ) to be slightly temperature dependent. The ratio of the second resistor (R2) 104 and the first resistor (R1) 103 is also known as a gain factor ( ^^) of the differential base-emitter voltage (∆ ^^^ா^, which simplifies the equation of the bandgap voltage ( ^^^ீ) to: ^^^ீൌ ^^^ா^ ^^ ∙ ∆ ^^^ாFIG. 2 shows a diagram of a quantizer 201 that uses the bandgap voltage ( ^^^ீ^ as a reference voltage ( ^^ோாி). The quantizer 201 converts an analog input voltage ^ ^^^^^ into a digital output code ^ ^^^௨௧) with the following equation: Where: ^^ is number of bits of the quantizer 201 output. Ideally, the bandgap voltage ^ ^^^ீ^ has a constant value (i.e., independent of temperature), the digital output code ^ ^^^௨௧) can then be related to the input voltage ^ ^^^^), i.e., the digital code has a value that represents the analog input voltage ^ ^^^^). Due to various error sources, the bandgap voltage ^ ^^^ீ) will not be a constant value, which causes deviation of the   digital output code ^ ^^^௨௧) from its ideal value. This deviation is known as the ADC gain error. The dominant error sources of the bandgap voltage ^ ^^^ீ) are the analog base-emitter voltage ( ^^^ா) and the gain factor ( ^^) which come from IC process variation of the BJTs or diodes, and resistors mismatch, respectively. The analog base-emitter voltage ^ ^^^ா) introduces the proportional to absolute temperature (PTAT) error (i.e., the error that increases with the absolute temperature) to the digital output code ( ^^^௨௧), which results in a temperature- dependent error. The variation in the gain factor ( ^^) depends on the variation of resistors matching and it can introduce a complex ADC error. FIG.3 shows a diagram of an ADC 300 according to the present invention. The ADC 300 comprises the quantizer 201, a switch 301, and a digital module 302. The ADC 300 has an analog input voltage terminal, an analog reference voltage terminal, and a digital output terminal. The reference input terminal of the ADC 300 is connected to the differential base- emitter voltage ^∆ ^^^ா^ node. The switch 301 can be configured to connect the input voltage terminal of the ADC 300 to either the analog input voltage ^ ^^^^) or the analog base-emitter voltage ( ^^^ா). The digital output terminal of the quantizer 201 is connected to the input terminal of the digital module 302. The digital module 302 comprises (a) a calculation unit 303, (b) memory units 304 and 305 and (c) non-volatile memory units 306 and 307. The digital module 302 uses at least 2 digital output codes from the quantizer 201 to perform a digital calculation to provide the digital output code ^ ^^^௨௧). The digital module 302 may use another digital code of temperature ( ^^்) to assist the digital calculation to provide the digital output code ^ ^^^௨௧). The ADC 300 converts analog signals to digital signals by using a two-step conversion method which can be described as follows. In the first step, two analog-to-digital conversions are performed by using at least one quantizer that uses the differential base-emitter voltage (ΔV^^^ as the reference voltage ^ ^^ோாி). In a first analog-to-digital conversion, the switch 301 is configured to connect the input terminal of the quantizer 201 to the analog input voltage ^ ^^^^) while the reference terminal of the quantizer 201 is connected to the differential base-emitter voltage ^ ^^ ^^^ா^. Thus, the digital output code of the quantizer 201 for the first conversion can be expressed by Where: N is the number of bits of the quantizer 201 In a second analog-to-digital conversion, the switch 301 is configured to connect the input terminal of the quantizer 201 to the analog base-emitter voltage ^ ^^^ா) while the reference terminal of the quantizer 201 is connected to the differential base-emitter voltage ( ^^ ^^^ா). The analog base-emitter voltage ^ ^^^ா^ may be generated from diodes-connected the BJTs or semiconductor junction diodes biased with a constant current source (not shown). Thus, the digital output code of the quantizer 201 for the second conversion can be expressed by Where:   N is the number of bits of the quantizer 201 The two digital output codes ^^^^and ^^^ாare respectively stored in the memory units 304 and 305 of the digital module 302. Note that the order of the two analog-to-digital conversions is not important. The analog input voltage ^ ^^^^^ may be converted before or after the analog base-emitter voltage ^ ^^^ா), or both the analog input voltage ^ ^^^^) and the analog base-emitter voltage ^ ^^^ா) may be converted simultaneously using two separate quantizers 201, with one dedicated quantizer 201 for each voltage. In the second step, the digital output code ^ ^^^௨௧^ is calculated by dividing the digital code ^ ^^^^^ obtained from the analog-to-digital conversion of the analog input voltage ( ^^^^^ by the summation of the digital code ^ ^^^ா^ obtained from the analog-to-digital conversion of the analog base-emitter voltage ^ ^^^ா^ and the gain factor ^ ^^). The digital output code ( ^^^௨௧) can be expressed by ^^^^ை^்ൌ ^^^^^ா ^ ^^^∗ ^2^ െ 1^Where ^^^ൌ ^^ ∗^2^െ 1^is the N-bit digital representation of the analog gain factor ^^. The above equation can be rearranged as follow: The above equation shows that the digital output code ( ^^^௨௧) of the ADC 300 is equivalent to the digital output code ^ ^^^௨௧) of a conventional ADC that uses the bandgap voltage ^ ^^^ீ) as the reference voltage ^^ோாி. However, the present method requires neither trimming of the bandgap voltage ^ ^^^ீ) nor trimming of the resistor ratio to obtain a high precision ADC conversion. By using the digital calculation performed by the digital module 302, the optimum of the resistor ratio or the gain factor ( ^^^ can be corrected and stored in the non-volatile memory unit 307 of the digital module 302. The parameters of the digital correction can easily be reconfigured via a memory programming to compensate for a batch- to-batch variation. However, the residue error of the bandgap voltage ^ ^^^ீ) is remained in ^^^ா, which can be corrected by using a correction factor ^ ^^^^. The correction factor ^ ^^^^ may be obtained by a calibration process and stored in the non-volatile memory unit 306 of the digital module 302. This will be explained by FIG.5. FIG.4 shows a flow chart of a method of converting analog signals to digital signals 400 using the ADC 300 according to the present invention. In the first step 401, the switch   301 is configured to connect the input voltage terminal of the quantizer 201 to a first analog input signal (either ^^^^or ^^^ா) and the quantizer 201 converts the first analog input signal ^either ^^^^or ^^^ா^ into a first digital code ^either ^^^^or ^^^ா, accordingly^. In the second step 402, the first digital code ^either ^^^^or ^^^ா^ is stored in one of the memory units (either 304 or 305) of the digital module 302. In the third step 403, the switch 301 is configured to connect the input voltage terminal of the quantizer 201 to a second analog input signal ^either ^^^^or ^^^ா^ into a second digital code ^either ^^^^or ^^^ா, accordingly^. In the fourth step 404, the quantizer 201 converts the second analog input signal ^either ^^^^or ^^^ா^ to a second digital code ^either ^^^^or ^^^ா, accordingly^. In the fifth step 405, the second digital code ^either ^^^^or ^^^ா^ is stored in one of the memory units (either 304 or 305) of the digital module 302. The first and second analog signals may be either ^^^^or ^^^ா, and the first analog signal is a different signal from the second analog signal. The first and second digital codes may be either ^^^^or ^^^ா, according to the first and second analog signals. The first digital code is a different code from the second digital code. The first and second digital codes are stored in two different memory units. Finally, in the sixth step 406, the output digital code ^ ^^^௨௧^ of the ADC 300 is calculated by dividing the digital code ^ ^^^^^ by the summation of the digital code ^ ^^^ா), and the gain factor ( ^^^. Note that, the order of the conversion is not important, the analog input voltage ( ^^^^^ may be converted before or after the analog base-emitter voltage ^ ^^^ா), or both the analog input voltage ( ^^^^^ a n d the analog base-emitter voltage ^ ^^^ா^ m a y b e c o n v e r t e d simultaneously using individual quantizer 201 for each voltage. A method of error correction is needed to compensate for errors in the bandgap voltage ^ ^^^ீ), which mainly comes from errors in the analog base-emitter voltage ^ ^^^ா^ and the gain factor ( ^^^. In the present invention, the gain factor ( ^^^ is implemented in the digital domain, thus, the error of the gain factor ^ ^^^ can be compensated by adjusting the digital calculation in the digital module 302. Therefore, the error in the analog base-emitter voltage ^ ^^^ா) is the remaining dominant source of error in the bandgap voltage ^ ^^^ீ^. The error of the analog base-emitter voltage ^ ^^^ா^ is largely proportional to PTAT which can be modelled as follow: ^^^ா ൌ ^^^ா,^ௗ ^ ^^^^^^^^^Where: ^^^ா,^ௗis the ideal analog base-emitter voltage ^ ^^^ா) characteristic without error. ^^^^^^^^^is the PTAT error voltage of the analog base-emitter voltage ^ ^^^ா). The error of the analog base-emitter voltage ^ ^^^ா) will affect the converted digital output code ( ^^^௨௧) as follows: Where: ^^^ா,^ௗis the ideal digital code value corresponding to the ideal analog base-emitter voltage ^ ^^^ா,^ௗ) without error.   ^^^ா,^^^is the error digital code corresponding to the PTAT error voltage ^ ^^^^^^ ^^^^ of the analog base-emitter voltage ( ^^^ா). Since the digital code ^ ^^^ா) is temperature dependent, the digital representation of the temperature ^ ^^்^ can be calculated by using the following equation: ^^ ൌ ^^ఈವ்^ಳಶାఈವ^ ^^ where A and B are constant parameters, and ^^^ൌ ^^ ∗ ^2^െ 1^ is the N-bit digital representation of the gain factor ^^. The digital code ^ ^^்^ represents the temperature in degree of Kelvin. The values of A, B, and ^^ depend on IC fabrication process and may be obtained from a curve fitting between the digital base-emitter voltage ^ ^^^ா) and the real temperature ^ ^^^^^^^ in degree of Kelvin. For an example embodiment and the process used in this invention, the values of A, B, and ^^ are approximately 650, -6, and 11.5, respectively. The digital code of temperature ^ ^^்) may represent the temperature in degree of Celsius by subtracting B with 273 (i.e., B = -279, in this case). By including the error from the analog base-emitter voltage ^ ^^^ா), the digital code of temperature ^ ^^்) can be expressed by ^^்ൌ ^^ఈವ^ಳಶ,^^ା^ಳಶ,^^^ାఈವ^ ^^ Next, the calibration temperature ^ ^^^^^) is measured by an external reference temperature sensor of the calibration system and can be expressed as follow: In the calibration, ^^^ா்^^^is equal to ^^^ா,^ௗthus the above equations of the digital code of temperature ^ ^^்) and the calibration temperature ^ ^^^^^) can be rearranged as follow: Since both ^^^^^^^^^and ∆ ^^^ாare proportional to absolute temperature (PTAT), the term^^^^^்^is approximately temperature independent. For simplicity, where ^^ is a correction factor which is a temperature-independent constant parameter, and it represents the error of the analog signal voltage ^ ^^^ா). Therefore, ^^^ாcan be compensated by subtracting with ^^^ா,^^^, thus, ^^^ா,^ௗcan be expressed as follow: ^^^ா,^ௗൌ ^^^ாെ ^^^ா,^^^   ൌ ^^^ாെ ^ ^^^∙ ^^^^Where ^^^ ൌ ^^ ∗^2^െ 1^is the digital representation of ^^.Therefore, the equation for calculating the digital output code ( ^^^௨௧^ can be modified to include the correction factor ^ ^^^^ as follow: The correction factor ^ ^^^^ can be easily added to compensate the error of the analog base-emitter voltage ^ ^^^ா^. In addition, since the digital code of temperature ^ ^^்^ represents the temperature in degree of Kelvin, this means that the converting analog signal voltage ^ ^^^ா) is also acts as a precision digital temperature sensor once calibrated by calculating the following equation: Therefore, when the ADC 300 converts analog signals to digital signals according to the method 400, the temperature at that point of time can be calculated. Moreover, the calibration temperature ^ ^^^^^^ is a single point of temperature, thus this is one point calibration that can calibrate the error over operating temperature range for both ADC 300 output and temperature output at the same time which significantly reduce the calibration cost. FIG.5 is a flow chart of a calibration method of the ADC 500 according to the present invention. In the first step 501, the analog base-emitter voltage ^ ^^^ா) is converted by the quantizer 201 which uses the differential base-emitter voltage (∆ ^^^ா) as the reference voltage ^ ^^ோாி^, wherein the analog base-emitter voltage ^ ^^^ா^ is generated from a BJT or a diode. The result of the said conversion is the digital output code ^^^ா்^^^. In the second step 602, the digital output code of calibration temperature ^ ^^்^^^) in Kelvin unit is calculated by using the following equation: In the third step 503, the calibration temperature ^ ^^^^^^ is measured by an external reference temperature sensor of a calibration system. In the fourth step 504, the correction factor ^ ^^^ is calculated by using the following equation: Finally, in the fifth step 505, the correction factor ( ^^^ is stored in the non-volatile memory unit 306 of the digital module 302. FIG 6. shows the effective result of reference voltage ( ^^^ீ) after trimming according to the invention. Since the present invention does not include an actual analog bandgap voltage to be the reference voltage of ADC, the bandgap voltage reference of the invention will be calculated by the bandgap voltage equation with the correction factor ^ ^^^ which results in the reduction of IC process variation.

Claims

1    Claims 1. A method for converting analog signals to digital signals comprising: Converting a first analog signal to a first digital code by using a quantizer; Storing the first digital code wherein the first digital code is stored in a digital module; Converting a second analog signal to a second digital code by using a quantizer; Storing the second digital code wherein the second digital code is stored in the digital module; Calculating a digital output code ( ^^^௨௧^, wherein the ^ ^^^௨௧^ represents one of the said analog signals.

2. A method for converting analog signals to digital signals of claim 1, wherein the first analog signal and the second analog signal can be an analog input voltage, either ^^^^or ^^^ா.

3. A method for converting analog signals to digital signals of claim 1, wherein converting the first analog signal and the second analog signal by using the quantizer further comprising steps of: Configuring a switch to connect the first analog signal to the input of the quantizer to obtain the first digital code, wherein the first analog signal can be either ^^^^or ^^^ா, and wherein the first digital code can be either ^^^^or ^^^ா, respectively; Configuring the switch to connect the second analog signal to the quantizer to obtain the second digital code, wherein the second analog signal can be either ^^^^or ^^^ாand is different from the first analog signal, and wherein the second digital code can be either ^^^^or ^^^ா, respectively, and is different from the first digital code.

4. A method for converting analog signals to digital signals of claim 1, wherein the first digital code and the second digital code can be converted simultaneously.

5. A method for converting analog signals to digital signals of claim 1, wherein the reference signal ^ ^^ோாி^ of the quantizer is a differential base-emitter voltage ^ ^^ ^^^ா).

6. A method for converting analog signals to digital signals of claim 5, wherein the differential base-emitter voltage ^ ^^ ^^^ா^ is derived from the junction voltages of bipolar junction transistors (BJTs) or diodes which are biased with different junction current densities.

7. A method for converting analog signals to digital signals of claim 1, where in the analog base-emitter voltage ^ ^^^ா^ is derived from the junction voltages of bipolar junction transistors (BJTs) or diodes.

8. A method for converting analog signals to digital signals of claim 1, wherein the digital code ^ ^^^௨௧^ can be calculated by using the first digital code together with the second digital code and with or without a correction factor ^ ^^^^.

9. A method for converting analog signals to digital signals of claim 8, where in the first digital code is a digital code ^^^^and the second digital code is a digital code ^^^ா.

10. A method for converting analog signals to digital signals of claim 8, where in the digital output code ^ ^^^௨௧^ expression is as follow: ^^ ^^^^ை^்ൌ ^^^^ா ^ ^^^∗ ^2 െ 1^

Citation Information

Patent Citations

  • Automatic reference voltage trimming technique

    US7433790B2

  • Methods and apparatus for higher-order correction of a bandgap voltage reference

    US7728575B1

  • Self-calibrating digital bandgap voltage and current reference

    US9323274B2

  • Device and method for absolute voltage measurement

    US20200099384A1