ADC Resolution Calculator

This tool calculates key performance metrics of an analog-to-digital converter: least significant bit (LSB) size, full-scale range, quantization error (±0.5 LSB), effective resolution (ENOB) considering noise, and theoretical signal-to-noise ratio (SNR). It also estimates whether the selected sample rate satisfies the Nyquist–Shannon criterion for the given signal bandwidth.

Use it during sensor interface design, data acquisition system specification, or when selecting an ADC for precision measurement applications (e.g., industrial sensing, medical instrumentation, or battery monitoring). It helps verify if your chosen ADC meets required dynamic range, noise floor, and sampling constraints before committing to hardware.

This parameter represents total input-referred noise—thermal, quantization, and circuit noise—expressed in root-mean-square LSBs. It directly impacts effective number of bits (ENOB); higher noise degrades resolution below the nominal bit count. Typical values range from 0.5 LSB (high-precision SAR ADCs) to >3 LSB (low-cost or high-speed delta-sigma converters).

“ADC Bits” is the ideal, manufacturer-specified resolution (e.g., 12-bit). “Effective Bits” (or ENOB) reflects real-world performance after accounting for noise, distortion, and nonlinearity—calculated as ENOB = log₂(SNR/6.02 + 1.76). A 12-bit ADC with 70 dB SNR yields ~11.3 ENOB, not 12.

Common V_ref values are 1.25 V (bandgap), 2.5 V, 3.3 V, or 5 V—often matching MCU supply rails or precision references. V_ref defines the ADC’s full-scale input range and directly sets LSB size (LSB = V_ref/2ⁿ). Using a noisy or unstable reference degrades accuracy more than bit count alone.

These inputs validate anti-aliasing compliance: per Nyquist, sample rate must exceed twice the signal bandwidth (fs > 2 × BW). The tool flags undersampling risk and helps determine minimum required oversampling ratio for noise shaping or digital filtering in delta-sigma systems.

Check for excessive front-end noise (e.g., unfiltered sensor output), poor PCB layout (ground loops, coupling), inadequate decoupling near V_ref/AVDD, or incorrect noise estimation. Also verify that “Noise (LSB RMS)” includes *all* noise sources—not just ADC intrinsic noise—but also amplifier, sensor, and layout contributions.

Yes—by comparing ENOB vs. sample rate trade-offs. For example, entering high noise + low fs favors delta-sigma (excellent noise shaping at low BW), while low noise + high fs points to SAR. The calculator highlights where oversampling benefits kick in and reveals resolution limits imposed by your analog chain.

Quantization error is the inherent uncertainty introduced when mapping a continuous analog voltage to a discrete digital code—bounded by ±0.5 LSB. It sets the fundamental limit on static accuracy; reducing it requires either more bits or a smaller V_ref, but only if system noise permits.

V_in determines the actual code value and relative position within the full-scale range. While not used in LSB or ENOB calculations directly, it helps assess headroom, clipping risk, and signal-to-quantization-noise ratio (SQNR) for that specific operating point—especially important for ratiometric or gain-calibrated systems.