ADC/DAC Resolution Calculator
LSB size, ENOB, and dynamic range
Required Parameters
Waiting for input data...
Quick Answer
LSB = Vref / 2^N. A 12-bit ADC with 3.3V reference has LSB = 0.806mV. Dynamic range = 6.02N + 1.76 dB.
Design Notes
More bits doesn't always mean more accuracy — noise floor, INL/DNL, and reference stability limit ENOB. A noisy 16-bit ADC may only achieve 12 ENOB. Use averaging (oversampling) to gain extra effective bits: 4× oversampling ≈ +1 bit. Always decouple the reference voltage with 100nF + 10µF.
Common Mistakes
- 1
Confusing resolution with accuracy — a 24-bit ADC with a noisy reference may only be accurate to 16 bits.
- 2
Ignoring input impedance requirements — SAR ADCs need low-impedance sources.
- 3
Not accounting for the ±0.5 LSB quantization error inherent in all ADCs.
Knowledge Base
What is ADC resolution?
Resolution is the number of discrete voltage levels an ADC can distinguish. An N-bit ADC has 2^N levels. 8-bit: 256 levels, 10-bit: 1024, 12-bit: 4096, 16-bit: 65536, 24-bit: 16.7 million. The LSB (least significant bit) voltage = Vref / 2^N. A 12-bit ADC with 3.3V reference resolves 0.806mV per step.
What is ENOB (Effective Number of Bits)?
ENOB represents actual performance including noise, distortion, and non-linearity: ENOB = (SINAD - 1.76) / 6.02. A 16-bit ADC with 90 dB SINAD has ENOB = (90-1.76)/6.02 = 14.7 bits. The remaining 1.3 bits are noise. ENOB is always less than the advertised resolution.
How do I improve ADC accuracy?
Hardware: (1) Clean reference voltage (use dedicated Vref IC, not VCC). (2) Decouple power pins with 100nF + 10µF. (3) Guard ring around analog input. (4) Keep digital signals away from analog. Software: (1) Oversampling: average 4× readings = +1 bit. (2) Calibration against known reference. (3) Digital filtering.
What is the difference between SAR, Delta-Sigma, and Flash ADCs?
SAR (Successive Approximation): 8-18 bits, 100ksps-5Msps, good general purpose. Delta-Sigma (ΔΣ): 16-32 bits, 10sps-1Msps, highest resolution, used for precision measurement. Flash: 4-8 bits, >1Gsps, fastest but lowest resolution, used in oscilloscopes and video. Pipeline ADC: 10-16 bits, 10-500Msps, bridges SAR and Flash.
How do I calculate quantization noise?
Quantization noise RMS = LSB / √12. The theoretical SNR of an ideal N-bit ADC = 6.02N + 1.76 dB. For a 12-bit ADC: SNR = 6.02(12) + 1.76 = 74 dB. This means the smallest signal distinguishable from quantization noise is 74 dB below full scale. In practice, thermal noise and INL/DNL further reduce this.
What voltage reference should I use?
Internal references: convenient but noisy (50-100ppm/°C). External references: LM4040 (simple shunt, ±0.1%), REF3033 (series, 8ppm/°C), ADR4533 (ultra-precision, 2ppm/°C). For 12+ bit ADCs, always use an external reference. Temperature coefficient matters most for precision: 10ppm/°C reference drifts 0.5mV/50°C on 3.3V.
Related Engineering Tools
555 Timer Calculator
Monostable and astable timing
I²C Pull-Up Resistor Calculator
Optimal pull-up for bus speed and capacitance
UART Baud Rate Error Calculator
Clock divider and baud rate accuracy
Crystal Oscillator Load Capacitor Calculator
External capacitor selection for crystal accuracy