What does ADC resolution mean?

ADC resolution is the number of bits N the converter uses to represent a sampled voltage. An N-bit ADC divides the full-scale range (FSR) into 2^N discrete codes. More bits mean finer voltage steps (smaller LSB) and a higher signal-to-noise ratio, but also higher cost, lower speed, and greater power consumption.

How is 1 LSB voltage calculated?

One LSB (least significant bit) equals the full-scale range divided by the number of codes: LSB = FSR / 2^N. For a 12-bit ADC with a 3.3 V reference the LSB is 3.3 V / 4096 = 805.66 µV. Any analog signal smaller than 1 LSB is indistinguishable from an adjacent code and is effectively lost.

What is quantisation noise and how is its RMS calculated?

Quantisation noise arises because the ADC rounds every sample to the nearest code. The rounding error is modelled as uniformly distributed between -LSB/2 and +LSB/2, giving an RMS value of LSB / sqrt(12). For the 12-bit, 3.3 V example above that is about 232 µV RMS.

What is the ideal ADC SNR formula?

The ideal signal-to-noise ratio for an N-bit ADC is SNR = 6.02 × N + 1.76 dB. The 6.02 factor comes from 20 × log10(2) and accounts for the 6 dB SNR improvement per additional bit. The 1.76 dB offset arises from the ratio of a full-scale sine wave RMS to the quantisation noise RMS. A 12-bit converter therefore has an ideal SNR of 6.02 × 12 + 1.76 = 74 dB.

What is ENOB and how do you compute it?

Effective Number of Bits (ENOB) tells you how many ideal bits a real ADC is equivalent to once non-linearities, thermal noise, aperture jitter, and other imperfections are included. It is calculated from the measured SNR: ENOB = (SNR_measured - 1.76) / 6.02. A 12-bit ADC typically achieves an ENOB of 10–11.5 bits depending on input frequency and temperature. Entering your datasheet SNR here gives you the true ENOB instantly.

What is the difference between dynamic range and SNR for an ADC?

Dynamic range (DR) is the ratio of the largest to the smallest signal the converter can faithfully represent: DR = N × 20 × log10(2) ≈ N × 6.02 dB. It describes the converter's amplitude window. Ideal SNR equals DR for a full-scale sine wave input, but SNR degrades with input level and frequency while DR is a fixed property of the bit depth. For most DC or low-frequency designs the two are effectively identical.

How many bits do I need for my application?

A useful rule of thumb is to add 2–4 bits to your required ENOB to allow for real-world degradation. Audio applications typically need 16–24 bits. Industrial sensors commonly use 12–16 bits. High-speed RF digitisers may only achieve 8–10 bits of real ENOB even at 14-bit resolution because jitter dominates at high frequencies. This calculator lets you test several bit depths side by side using the comparison table.

ADC Resolution Calculator

An ADC resolution calculator that tells you everything you need to know about an analog-to-digital converter before you commit to a part in your schematic. Enter the bit depth (1–32 bits), the reference voltage range, and — optionally — the signal bandwidth and measured SNR from a datasheet; the tool instantly computes the LSB voltage, quantisation noise, ideal SNR, effective number of bits (ENOB), and dynamic range, with a full step-by-step working log and a side-by-side comparison table for common ADC widths.

How it works

Every analog-to-digital converter maps a continuous voltage to one of 2^N discrete integer codes. The four fundamental performance metrics all follow from this single relationship:

Full-Scale Range (FSR) is simply the span of voltages the ADC can represent: FSR = V_ref_high - V_ref_low. For a unipolar 3.3 V supply that is 3.3 V; for a bipolar ±2.5 V reference it is 5.0 V.

1 LSB is the voltage step between adjacent codes: LSB = FSR / 2^N. It is the finest distinction the converter can make. Any input change smaller than half an LSB will not change the output code at all.

Quantisation noise is the irreducible RMS error from rounding each sample to the nearest code. Assuming the input signal is busy enough to randomise the error uniformly across the ±LSB/2 window, the noise power is LSB^2/12 and the RMS noise = LSB / sqrt(12).

Ideal SNR for a full-scale sine wave input is the textbook result: SNR = 6.02 × N + 1.76 dB. Every additional bit adds almost exactly 6 dB.

ENOB (Effective Number of Bits) converts a measured SNR back into bits: ENOB = (SNR_dB - 1.76) / 6.02. Leave the SNR field blank for the ideal figure; fill it in with a datasheet number to see the real ENOB.

Nyquist minimum sample rate is the theoretical floor set by the Shannon-Nyquist theorem: f_s_min = 2 × BW. In practice, engineers use an oversampling ratio of 2–10x to ease the anti-aliasing filter roll-off requirements.

Worked example

A 12-bit SAR ADC referenced to 0 V / 3.3 V:

Number of codes = 2^12 = 4,096
FSR = 3.3 V - 0 V = 3.3 V
1 LSB = 3.3 V / 4096 = 805.66 µV
Quantisation noise RMS = 805.66 µV / sqrt(12) = 232.6 µV RMS
Ideal SNR = 6.02 × 12 + 1.76 = 74.0 dB
Dynamic range = 12 × 6.0206 = 72.25 dB
Ideal ENOB = (74.0 - 1.76) / 6.02 = 12.00 bits

If the datasheet lists a measured SNR of 70.4 dB at 10 kHz input:

ENOB = (70.4 - 1.76) / 6.02 = 11.40 bits (the lost 0.6 bits comes from real-world noise and distortion)

Bits	Codes	LSB (3.3 V)	Ideal SNR
8	256	12.89 mV	49.92 dB
12	4,096	805.66 µV	74.0 dB
16	65,536	50.35 µV	98.1 dB
24	16,777,216	196.7 nV	146.2 dB

Formula reference

Codes = 2^N
FSR = V_ref_high - V_ref_low
LSB = FSR / 2^N
Q_noise_RMS = LSB / sqrt(12)
SNR_ideal = 6.02 × N + 1.76 (dB)
Dynamic range = N × 20 × log10(2) ≈ N × 6.0206 (dB)
ENOB = (SNR_dB - 1.76) / 6.02
f_s_min = 2 × BW (Nyquist)

All calculations are performed in your browser — no data is sent to any server.