24/7 emergencies 814 777 6921812 230 6079WhatsApp 811 786 0768
Electrónica MonterreyElectrónica MonterreyLaboratorios Creativeit Talk to an engineer
Menu

How do you use the average function to reduce noise?

The average function sums many acquisitions of a repetitive signal and calculates their mean; uncorrelated random noise is reduced, and the signal that repeats on every trigger remains.

Why averaging reduces noise

Each time the oscilloscope triggers, it captures the signal plus a different sample of noise. If the noise is random and unrelated to the trigger, its positive and negative excursions tend to cancel out when averaged, while the deterministic part of the signal is identical in every capture and is preserved. The improvement in signal-to-noise ratio grows with the square root of the number of acquisitions: quadrupling the average reduces the noise to approximately half.

How to configure it

  1. Obtain a stable trigger on the repetitive signal. Without a coherent trigger, each capture is misaligned and averaging erases the signal itself.
  2. Choose the “average” acquisition mode and set the number of acquisitions.
  3. Wait for the trace to stabilize. The result improves gradually, and each increment contributes less than the one before.
  4. If the instrument allows it, use continuous averaging when you need to see slow changes, and block averaging when you want a final result.

What averaging cannot do

  • It does not remove synchronous noise. Interference that repeats on every trigger, such as coupling from a clock or from power supply switching, is averaged just like the signal and remains.
  • It is no use for single events. It requires repetition.
  • It hides real variations. Jitter, amplitude changes and intermittent faults are smoothed out and can disappear from the screen.
  • It slows the update. The instrument needs all the acquisitions before it can show a stable result.

Averaging versus high resolution

High-resolution mode averages neighboring samples within a single acquisition. It works with non-repetitive signals and gains vertical resolution, but in exchange it limits the effective bandwidth and smooths fast edges. Waveform averaging preserves bandwidth but demands repetition.

How to interpret the result

If the noise drops and the shape remains, the noise was random. If a stable ripple is still there after averaging, it is not random noise: look for a source synchronized with the signal or a measurement artifact. Always compare with the unaveraged trace before accepting a measurement, because averaging can make a signal look cleaner than it really is.

We do real tests, on video! Watch them on our YouTube channel →

Do you have a problem nobody has been able to measure?

Talk directly to the engineer in charge, with no middlemen.

Talk to an engineer