Two of the most important parameters of the scope in the hardware test

1. Oscilloscope bandwidth

The maximum frequency when the amplitude of sine wave measured by oscilloscope is no less than the amplitude of real sine wave signal 3dB (power half of the original). If a waveform with fixed amplitude is input and the signal frequency is increased, the position of -3db is 0.707 times the actual amplitude of the oscilloscope voltage.
When the frequency of the measured sine wave is equal to the bandwidth of the oscilloscope (the response of the oscilloscope's amplifier is a first order gaussian), the amplitude measurement error is about 30%. If you want to measure a sinusoidal wave with an amplitude error of only 3%, the frequency of the measured sinusoidal wave is much lower than the bandwidth of the oscilloscope (about 0.3 times the bandwidth of the oscilloscope). The general rule of thumb for measuring a signal with an oscilloscope is that the bandwidth of the oscilloscope is five times the frequency of the signal being measured. For example, a 66MHz clock signal requires a 330MHz bandwidth oscilloscope.
How do you get 5 times the frequency? The signal bandwidth has nothing to do with the rate, only with the edge. When the signal is 7% of the period, the signal can be effectively restored, and the oscilloscope bandwidth must be ≥0.35/Tedge, so the oscilloscope bandwidth must be ≥5*F.





2. Sampling rate

By measuring the voltage amplitude of the waveform of equal time interval and converting the voltage into digital information represented by 8-bit binary code, this is the sampling of digital oscilloscope. The smaller the time interval between the two samples, the closer the reconstructed waveform is to the original signal. The Sampling Rate is the reciprocal of the Sampling interval. For example, if the sampling rate of an oscilloscope is 10G times per second (10GSa/s), that means sampling every 100ps.
According to Nyquist sampling theorem, for sinusoidal waves, at least two or more samples are needed in each cycle to ensure that the digitized pulse train can accurately restore the original waveform. If the sampling rate is lower than Nyquist sampling rate, it will result in Aliasing.
The unit of sampling rate is MS/s (MegaSamples per second) or GS/s(GigaSamples per second). Generally, the sampling rate parameters published by each oscilloscope refer to the highest sampling rate of a single channel. If a two-channel oscilloscope is used, the sample rate parameter published is 1GS/s. When two channels are used simultaneously, the maximum sample rate of each channel is 500MS/s.
There is a rule of thumb for comparing sampling rates with signal bandwidths: if the oscilloscope is interpolated (to be filtered for re-generation between sampling points), the ratio should be at least 4:1; If there is no sinusoidal interpolation, the ratio of 10:1 should be taken.

These are the two most important parameters of an oscilloscope.