Skip to Content

dsPIC33A 12-bit 40MSPS Analog-to-Digital Converter (ADC) Deep Dive

Oversampling: Throughput

Last modified by Microchip on 2026/03/31 11:32
   Throughput  Oversampling Metrics   

Click Play ► to listen to the commentary.

The Analog-to-Digital Converter (ADC) clock (TAD) can operate at 80 MHz, or 12.5 ns.

When using oversampling, always apply input filtering to prevent aliasing and ensure accurate conversions.

For oversampling, use Trigger 1 to start the initial conversion and Trigger 2 for subsequent conversions. When ACCBRST is set to 1, oversampling proceeds automatically and no additional triggers are needed.

oversampling table

The data in the table presents key oversampling operational metrics to take note of, from Exact Number of Bits (ENOB) to throughput.

Oversampling is most effective when there is enough signal noise to dither the output around 1 Least Significant Bit (LSB).

Using 4x to 16x oversampling can add 1–2 bits of resolution and help reduce device noise.

Higher oversampling rates, such as 64x or 256x, can further improve resolution and minimize noise from the application circuit.