Scope Measurement Hints

How to prevent your scope from aliasing

Here are a few things you can do to help ensure that your oscilloscope waveform is not aliased:

1. Check for a stable trigger. Aliased waveforms sometimes appear to drift across the oscilloscope screen or look "untriggered".

2. Make sure the scopeís effective sample rate is as fast as possible. Most oscilloscopes decrease their effective sample rate as the time/div is slowed. (Note that repetitive scopes can have a high effective sample rate, even with a low actual sample rate). If you suspect an aliased waveform, increase the sweepspeed to ensure that the scope is sampling as fast as possible for proper display of the waveform. For example, Figure 1 shows an aliased waveform at 5 ms/div. The actual signal is 13 MHz but the scope measures 50 Hz! At this sweepspeed the scope is sampling at 10 kSa/s, severely violating the Nyquist* criteria. Figure 2 shows the same signal at 20ns/div. At this sweepspeed the scope is sampling at 2GSa/s. The scope displays the correct signal and measures the correct frequency, 13 MHz.

3. Some oscilloscopes employ techniques to prevent aliasing. HP 54600- series oscilloscopes use a patented proprietary algorithm to reduce aliasing. In Figure 1 the signal is aliased. Figure 3 shows the same signal on an HP 54600B oscilloscope. The waveform is not aliased at the same sweepspeed due to the anti-aliasing algorithms.

4. Another method for determining if aliasing is occurring is to put the scope in peak detect mode. Peak detect maintains the maximum sample rate and plots the max/min values on display. Using peak detect, no "alias" will appear.

* The Nyquist sampling theorem states that for a baseband signal to be faithfully reproduced in sampled form, the sample rate must be greater than twice the highest frequency present in the signal.