Bandwidth limiting and averaging can be useful tools to reduce the amount of noise in a measurement, but must be used with caution. Both of these tools have advantages if used appropriately, but can give false impressions if used without consideration of the measurement being taken.

Although bandwidth limiting significantly reduces high-frequency noise present in a waveform, bandwidth limiting can also attenuate or filter components of the waveform that are important to see for debug, such as overshoot and undershoot. This can lead to inaccurate measurement readings, as seen in Figure 3-13 and Figure 3-14. If there are no expected high-frequency components in the intended measurement, bandwidth limiting can be used to get a clearer picture of the electrical behavior of the circuit with less noise, but in some cases there are unexpected high-frequency signals due to noise coupling or resonances from underdamped filters or parasitics. We recommend starting with full bandwidth and only limit the bandwidth after some confidence has been developed that the measurement does not have important high-frequency components.

Averaging can be a useful tool when there is no easy way to reduce noise through long measurement wires. This method requires that the waveform to be measured is consistent between measurements. Figure 3-16 shows how the noise in the measurement shown in Figure 3-15 is reduced by averaging 5 samples on the load transient shown on Channel 3.