ESP32 ADC Guide: Advanced Functions, Accuracy & Calibration

ADCs convert analog signals (like voice waves) into digital signals for microcontroller processing. Resolution depends on bit depth: 8-bit ADCs provide 0-255 codes, while 10-bit ADCs provide 0-1023 codes. Higher bit depth means better resolution (smaller detectable voltage changes). The Nyquist-Shannon sampling theorem states sampling frequency must be at least twice the signal frequency, but practical applications require 4-8 times the signal frequency for accurate reconstruction. ADCs have positive and negative reference voltages that define the input range. Resolution equals maximum code divided by reference voltage range. Lower reference voltages provide better resolution for small signals, while higher voltages allow measuring larger signals.

An Analog-to-Digital Converter (ADC) converts continuous-time, continuous-amplitude analog signals (like sensor outputs) into discrete-time, discrete-amplitude digital signals. Three critical parameters define ADC performance: resolution, quantization error, and sampling rate. Resolution determines the number of discrete output levels (N bits), calculated as Output = (Input Voltage / Vref) × 2^N. Quantization error arises because continuous analog inputs have infinite values while ADC outputs are finite, creating unavoidable rounding errors. The maximum quantization error is 0.5 LSB, uniformly distributed between -0.5 LSB and +0.5 LSB. Understanding these fundamentals is essential for designing effective data acquisition systems.

An ADC (Analog-to-Digital Converter) converts analog signals to digital binary numbers. Key concepts include: (1) Resolution (n-bit) determines the number of possible digital values (2^n), with common values being 8, 10, 12, or 16 bits; (2) Step size is the smallest voltage change an ADC can detect, calculated as reference voltage divided by number of steps; (3) Higher resolution means smaller step size and better precision; (4) Conversion time is the duration to complete one conversion; (5) ADCs can have serial or parallel output configurations. For example, an 8-bit ADC with 5V reference has 256 steps and 19.53mV step size, while a 16-bit ADC has 65,536 steps and 0.076mV step size.

An Analog-to-Digital Converter (ADC) transforms continuous analog signals into discrete digital values (zeros and ones). The reference voltage is critical for accurate conversion, serving as the basis for calculating results. Reference voltages can come from internal sources or the power supply, but power supply noise causes measurement errors. Dedicated reference voltage circuits with regulators (0.5-1% error) outperform standard power regulators (±5% error). Resolution is calculated as (Vref_positive - Vref_negative) / 2^n, where n is the bit count. For 5V reference and 8-bit resolution, each count represents approximately 19.607 mV.

An Analog-to-Digital Converter (ADC) converts continuous analog signals into discrete digital values through three sequential processes: sampling (capturing instantaneous signal values at regular intervals following the Nyquist theorem requiring sampling rate ≥ 2× maximum signal frequency), quantization (mapping infinite amplitude values to finite discrete levels), and encoding (converting quantized values into binary digital codes); the resolution of an ADC, defined by the number of bits (n), determines its ability to detect minimum input changes, with higher bit counts providing finer resolution (e.g., an 8-bit ADC with 10V full-scale range has a resolution of 39mV, while a 3-bit ADC has 1.25V resolution).

The ESP32 is a powerful microcontroller with built-in Wi-Fi and Bluetooth connectivity, faster than Arduino at up to 240 MHz, featuring over 30 GPIO pins for device control and sensor reading. To program it, download and install the Arduino IDE, add the ESP32 board manager URL, install the ESP32 package, configure the board and port settings, and write a simple sketch using setup() for initialization and loop() for continuous operations, such as controlling the built-in blue LED connected to GPIO pin 2 using digitalWrite() commands.

Programming the ESP32 using Arduino IDE requires installing the ESP32 Arduino core from GitHub. Users download the ZIP file, extract it to the Arduino hardware folder structure (Arduino → hardware → espressif → esp32), then run the get.exe tool to complete installation. In the Arduino IDE, select 'ESP32 Dev Module' from Tools → Board, set upload speed to 115200 baud, and choose a programmer. The blink example demonstrates basic operation: code toggles the built-in LED every 500ms. Connect the board, press Ctrl+U to upload, and verify successful operation by observing the blinking LED.

The ESP32 is a dual-core 32-bit microcontroller module manufactured by Expressif Systems, offering ten times the performance of the Arduino Uno (8-bit vs 32-bit architecture). Key features include integrated Wi-Fi, Bluetooth, and Bluetooth Low Energy capabilities, making it ideal for IoT applications. The device supports multiple communication protocols including SPI, I2C, and I2S buses, with up to 18 12-bit ADCs, two 8-bit DACs, and 10 capacitive touch sensors. Most pins have multiple functions with potential conflicts preventing simultaneous use. To program the ESP32 using Arduino IDE, users must add board support via the Boards Manager by visiting Expressif's GitHub page, copying the JSON file URL, pasting it into Arduino preferences under Additional Board Managers URLs, then installing via Tools > Board > Boards Manager. Once installed, ESP32 boards appear in the board selection menu.

To program an ESP32 microcontroller using Arduino IDE, first connect the ESP32 Dev Kit to your computer via USB cable, then add ESP32 board support by configuring the Additional Board Manager URLs in Arduino IDE preferences with the official Espressif URL and an additional forum-provided URL, install the ESP32 boards package through the Board Manager, select the appropriate ESP32 board and COM port in the Tools menu, and finally upload your Arduino sketch to blink the built-in LED connected to IO2 pin or an external LED connected to D2 pin.

The Arduino IDE is the most accessible environment for ESP32 programming, supporting Windows, Linux, and Mac with extensive examples and community support. Installation involves downloading from the official Arduino website, running the installer, and accepting the agreement. Board support is added through Tools > Board > Board Manager by searching for 'ESP32' and installing the Espressif Systems board. Modern operating systems typically include pre-installed drivers, but users may need to download drivers from manufacturers if the board isn't recognized. Basic programming uses setup() for one-time initialization and loop() for repeated execution. The Serial Monitor displays program output and allows communication with the microcontroller. A blinking LED circuit requires an LED and 220 ohm resistor connected to a GPIO pin configured as output. Analog reading uses analogRead() with the 12-bit ADC, producing values from 0 to 4095 corresponding to 0V to 3.3V. Digital button reading uses digitalRead() after configuring pins as INPUT_PULLUP, which provides internal pull-up resistors keeping pins at high logic level when not pressed. PWM controls LED brightness by varying duty cycle, with analogWrite() accepting values from 0 to 255 producing voltages from 0V to 3.3V.

This segment covers the mathematical techniques for solving linear equations and fitting polynomial curves to data. The instructor demonstrates step-by-step methods for algebraic manipulation and equation solving, which form the foundation for regression analysis. The lecture explains how polynomial regression works by finding the best-fitting curve through data points, minimizing the distance between the model and actual observations.

Curve fitting converts random data points into mathematical equations. Two main equation types exist: linear (y = a + bx) for straight lines, and quadratic/parabolic (y = a + bx + cx²) for curved lines. For linear equations with two unknowns (a and b), multiply by Σ and Σx to derive two equations. For quadratic equations with three unknowns (a, b, c), multiply by Σ, Σx, and Σx² to derive three equations. The complete procedure involves: determining the equation type, deriving equations, creating a data table with necessary columns, calculating sums, substituting values, solving the system, and substituting back to get the final equation.

Curve fitting is a mathematical process that finds a polynomial function passing exactly through a given set of data points; for n points, a unique polynomial of degree (n-1) exists, which can be determined by solving a system of linear equations using methods like Cramer's rule, as demonstrated by finding the quadratic equation y = 2x² - 3x + 4 that passes through the points (4,24), (2,6), and (-1,-19).

Polynomial curve fitting determines the best-fit polynomial equation for a given set of data points. The key principle is that the degree of the polynomial is always one less than the number of data points: two points fit a first-degree (linear) polynomial, three points fit a second-degree (quadratic) polynomial, and so on. Each polynomial degree has a specific general form: linear is f(x) = ax + b, quadratic is f(x) = ax² + bx + c, cubic is f(x) = ax³ + bx² + cx + d, and fourth-degree is f(x) = ax⁴ + bx³ + cx² + dx + e. This systematic approach allows for precise interpolation when the number of data points matches the polynomial's degree plus one.

Curve fitting is the process of finding the best fit curve for a given set of data using algebraic equations. Linear curve fitting involves fitting a straight line to data, which can be represented as y = a + bx (y as dependent variable) or x = a + by (x as dependent variable). The normal equations for y = a + bx are: Σy = na + bΣx and Σxy = aΣx + bΣx². For x = a + by, the equations are: Σx = na + bΣy and Σxy = aΣy + bΣy². To perform curve fitting, create a table with columns for x, y, x², and xy values, then calculate the sums (Σ) of each column.