DAC Explained: Converting Digital Values to Voltage

What Is a DAC?

A Digital-to-Analogue Converter (DAC) is the opposite of an ADC. Where an ADC reads a voltage and converts it to a digital number, a DAC takes a digital number and produces a proportional analogue voltage. DACs are essential wherever a digital system must interact with the analogue world:

Audio playback — MP3 files are digital data; your speakers need an analogue signal
Waveform generation — producing sine waves, ramps, or arbitrary signals from a microcontroller
Motor control — setting a voltage reference for a motor driver
Calibration outputs — generating precise reference voltages in test equipment
Display brightness control — producing an analogue voltage to drive a backlight

How a DAC Works

A DAC converts an n-bit digital input code into an analogue voltage according to:

Vout = (Code / 2ⁿ) × Vref

Where:

Code is the input value (0 to 2ⁿ − 1)
n is the number of bits (resolution)
Vref is the full-scale reference voltage

Worked Example: 8-bit DAC, Vref = 5V

Maximum code = 2⁸ − 1 = 255. Input code = 128:

Vout = (128 / 256) × 5V = 0.5 × 5V = 2.5V

Exactly half the reference voltage. Code 0 outputs 0V; code 255 outputs:

Vout = (255 / 256) × 5V ≈ 4.98V

Note: a DAC can never output exactly Vref — the maximum output is one LSB below it.

Worked Example: 12-bit DAC, Vref = 3.3V

Code = 2048 (mid-scale):

Vout = (2048 / 4096) × 3.3V = 0.5 × 3.3V = 1.65V

LSB Voltage: The Smallest Step

The voltage change per code increment (Least Significant Bit value) is:

LSB = Vref / 2ⁿ

This is the smallest voltage change the DAC can produce.

Resolution	Vref	LSB size
8-bit	5V	19.53 mV
10-bit	3.3V	3.22 mV
12-bit	3.3V	0.806 mV
16-bit	5V	76.3 µV

Higher resolution allows smoother signal generation. For audio, 16-bit at 44.1kHz (CD quality) gives 65,536 voltage steps — smooth enough that step transitions are inaudible.

DAC Resolution: How Many Bits Do You Need?

The required resolution depends on the application:

PWM substitute (servo, LED dimming): 8–10 bits is usually enough
Audio output: 16-bit minimum for CD quality; 24-bit for studio work
Signal generation (function generator): 12–16 bits for clean waveforms
Precision voltage reference: 12–16 bits, often with external Vref
Calibration outputs: 16–24 bits for lab-grade accuracy

Types of DAC Architecture

R-2R Ladder DAC

Uses a resistor ladder network where each bit switches a resistor in or out. Inexpensive, easy to implement with discrete components, and widely used in low-to-medium resolution applications.

String DAC (Resistor String)

A chain of equal resistors forms a voltage divider; a multiplexer selects the tap point. Used in LCD drivers and lower-resolution applications where monotonicity is critical.

Sigma-Delta (ΔΣ) DAC

Oversamples at a much higher rate than required and uses noise shaping to achieve very high resolution. Used in audio DACs (PCM5102, CS4398) and precision instruments. High resolution (24-bit+) but lower maximum frequency.

Current Steering DAC

Switches binary-weighted current sources; the output current is converted to a voltage via a load resistor. Fast, used in RF applications and high-speed signal generation.

Common DAC Chips

Chip	Bits	Interface	Vref	Notes
MCP4725	12-bit	I²C	Internal/external	Popular, low cost, non-volatile register
DAC7571	12-bit	I²C	External	Texas Instruments, rail-to-rail
MCP4921	12-bit	SPI	External	Single channel, buffered output
PCM5102	32-bit	I²S	Internal	Audio DAC, used in HiFi projects
DAC8552	16-bit	SPI	External	Dual channel, precision
AD5685	16-bit	SPI	Internal/external	Quad channel, Analog Devices

DAC vs PWM: When to Use Which

Many microcontrollers use PWM (Pulse-Width Modulation) as a low-cost substitute for a true DAC. A filtered PWM signal approximates a DC voltage, but with limitations:

Feature	True DAC	PWM + filter
Speed	Fast (µs)	Slow (requires RC filter settling time)
Noise	Low (no switching)	Switching noise from PWM frequency
Resolution	12–24 bit	8–16 bit (limited by timer)
Cost	Requires external IC	Free (built into MCU)
Complexity	Simple to use	Requires filter design

Use a true DAC when you need a stable, low-noise voltage or audio-quality output. PWM is fine for motor speed control, LED brightness, or servo positioning.

Practical Tips

Buffer the DAC output — most DACs have a limited output current. Use an op-amp voltage follower to drive low-impedance loads.
Decouple the supply — place 100nF and 10µF capacitors from VCC to GND close to the DAC.
Use a precision external reference for critical applications — the internal Vref can drift ±0.1% or more with temperature.
Account for glitches — when the input code changes, transient spikes can occur as internal switches transition. Deglitch circuits or sample-and-hold stages are used in professional systems.
Check output settling time in the datasheet — this limits the maximum waveform frequency you can generate.

Use the DAC Calculator

The DAC Calculator on DevGizmo lets you enter bit depth, reference voltage, and input code to instantly calculate output voltage and LSB size. Presets for popular chips like the MCP4725 and DAC8552 are included.