✦ScienexySCIENCE KNOWLEDGE LIBRARY
General science · SCI-FORM-ENG-1541

ADC Quantization Step Size

The ADC quantization step size is the approximate input-voltage interval represented by one digital code. For an N-bit ideal analog-to-digital converter, the input range is divided into 2ᴺ quantization intervals.

\(\Delta = (V_{\max} - V_{\mathrm{min}})/2N\)

What this formula means

The ADC quantization step size is the approximate input-voltage interval represented by one digital code. For an N-bit ideal analog-to-digital converter, the input range is divided into 2ᴺ quantization intervals.

Scientific principle

Principle

An analog-to-digital converter converts a continuous input voltage into a finite set of digital codes. Since only a limited number of codes are available, nearby analog values are represented by the same code, creating quantization.

Derivation

  • 1. An N-bit ADC can represent 2ᴺ distinct digital codes.

  • 2. Let the input voltage range extend from V_min to V_max.

  • 3. The total input range is V_FS = V_max − V_min.

  • 4. Dividing the input range into 2ᴺ quantization intervals gives the nominal step size.

  • 5. Therefore, Δ = (V_max − V_min)/2ᴺ.

  • 6. For an ideal uniform quantizer with rounding, the maximum quantization error is approximately ±Δ/2, excluding overload and endpoint conventions.

Final Result

Δ = V_FS/2ᴺ

Quantization Error

−Δ/2 ≤ e_q ≤ Δ/2

Resolution Relation

Number of quantization levels = 2ᴺ

Variables and their meanings

SymbolMeaningUnit

Δ

Quantization step size or nominal LSB interval

V

V_FS

Full-scale input range, V_max − V_min

V

V_max

Maximum input-range voltage

V

V_min

Minimum input-range voltage

V

N

ADC resolution in bits

bit

e_q

Quantization error

V

SI units and conversions

Step Size

V

Input Range

V

Resolution

bit

Dimensional Formula

Voltage: [M L² T⁻³ I⁻¹]

Conditions of use

  • The ADC is assumed to have N-bit resolution and 2ᴺ quantization levels.

  • The converter is assumed to use uniform quantization.

  • V_FS must represent the actual configured analog input range.

  • The formula assumes ideal quantization and does not include gain, offset, or nonlinearity errors.

  • The distinction between quantization intervals and the difference between endpoint code values must be maintained.

Limitations and assumptions

  • Some ADC specifications define code spacing using endpoint conventions involving 2ᴺ − 1; the exact convention should be checked in the datasheet.

  • The formula does not include thermal noise, reference-voltage noise, or sampling noise.

  • Actual effective resolution can be lower than nominal bit resolution.

  • Input voltages outside the permitted range can cause clipping or overload.

  • The quantization-error bound assumes ideal rounding; truncation can produce a different error interval.

Step-by-step worked examples

ProblemSolutionAnswer

An ideal 12-bit ADC has an input range from 0 V to 5 V. Calculate its nominal quantization step size.

  • Given: N = 12, V_max = 5 V, V_min = 0 V.

  • Number of quantization levels = 2¹² = 4096.

  • Input range: V_FS = 5 − 0 = 5 V.

  • Use: Δ = V_FS/2ᴺ.

  • Δ = 5/4096.

  • Δ ≈ 0.0012207 V.

The nominal quantization step size is approximately 1.221 mV.

An ideal 10-bit ADC operates over a 0–3.3 V input range. Determine its quantization step size and ideal rounding-error bound.

  • Given: N = 10 and V_FS = 3.3 V.

  • Number of levels = 2¹⁰ = 1024.

  • Step size: Δ = 3.3/1024.

  • Δ ≈ 0.0032227 V = 3.223 mV.

  • Maximum ideal rounding error = Δ/2.

  • Maximum error ≈ ±1.611 mV.

The nominal step size is approximately 3.223 mV, with an ideal rounding-error bound of ±1.611 mV.

Real-world applications

  • Digital sensor interfaces

  • Data-acquisition systems

  • Audio recording and processing

  • Microcontroller analog-input systems

  • Measurement-instrument resolution analysis

Formula Meaning

The ADC quantization step size is the approximate input-voltage interval represented by one digital code. For an N-bit ideal analog-to-digital converter, the input range is divided into 2ᴺ quantization intervals.

Sources and further reading

No source has been supplied for this item yet. Treat the claim as needing editorial verification.

← Back to Formula Hub · Suggest a correction or source