ADC : Offset Error and Gain Error Explained

ADC Performance Parameters: Beyond Basics
📌 Basic ADC parameters like resolution, reference voltage, and sampling rate are insufficient for optimal application selection.
🧐 A 12-bit ADC with an Integral Non-Linearity (INL) error of 4 Least Significant Bits (LSB) effectively yields only 10 bits of accuracy.
⚙️ For accurate ADC selection, parameters related to DC errors, specifically gain error and offset error, must be analyzed.

Understanding Offset Error
↔️ Offset error occurs when the ADC's transfer function shifts horizontally (left or right) relative to the ideal transfer function.
🛑 Applying 0 Volts input and observing an output code other than all zeros indicates the presence of an offset error.
📈 Positive offset error occurs when the actual transfer curve lies above the ideal curve; negative offset error occurs when it lies below.
📊 Offset error can be quantified in LSB, volts, or as a percentage of the full-scale voltage. For an N-bit ADC with reference voltage $V_{REF}$, the voltage error is $\text{Error}_{LSB} \times \frac{V_{REF}}{2^N}$.

Understanding Gain Error
📉 Gain error represents the difference in the slope between the ideal and actual transfer functions, measured after calibrating for offset error.
➕ A positive gain error means the actual slope lies above the ideal slope; a negative gain error means the actual slope lies below the ideal slope.
⚠️ Both offset and gain errors reduce the ADC's overall conversion range.

Key Points & Insights
➡️ Even if required accuracy is met by resolution (e.g., needing 10 bits, using 12 bits), non-linearity errors like INL can degrade effective accuracy.
➡️ Offset error calibration using software can reduce conversion range; for a 12-bit ADC, an offset of $12\text{ LSB}$ translates to approximately $14.65\text{ mV}$ error relative to $V_{REF}$.
➡️ Gain error is quantified by comparing the actual voltage producing the full-scale output code against the ideal voltage, reflecting the difference in slope.

📸 Video summarized with SummaryTube.com on Dec 02, 2025, 03:32 UTC