How to Test the Refresh Rate of an Oscilloscope

Waveform refresh rate is an important indicator of an oscilloscope. Together with the sampling rate, it directly reflects the oscilloscope's ability to capture waveform details. Currently, mainstream oscilloscopes on the market all exceed 10,000 wfms/s (typical value). However, this indicator is generally given by oscilloscope manufacturers, and few articles mention how to test this indicator, resulting in users being unable to truly measure it. This article will introduce a simple method for roughly estimating the refresh rate of an oscilloscope.

2. Test ideas

The mode of each acquisition of the oscilloscope is shown in Figure 1 below.

The time to collect a waveform is divided into two parts. The first part is the sampling time, which is affected by the sampling rate and storage depth; the second part is the dead time, which is used by the processor to calculate waveform data, display, measure, etc. The two parts together determine the refresh rate, so the refresh rate of a long-storage oscilloscope must be lower than that of a short-storage oscilloscope, which also shows that in order to pursue a high refresh rate, the storage depth during RUN becomes lower.

When the waveform falls within the sampling time, the oscilloscope can trigger to the edge under appropriate settings, which is the stable trigger we usually see. If the waveform falls within the sampling time, the oscilloscope cannot display it. Therefore, consider using two pulses to observe the oscilloscope trigger to calculate the refresh rate.

As shown in the figure below, when two pulses are displayed simultaneously within one sampling time, the oscilloscope can stably trigger on the first rising edge, and the second edge is also captured and may be displayed on the screen or off the screen.

When the second pulse is delayed to the dead zone, the oscilloscope can also stably trigger on the first rising edge, but the second edge is not collected, and the user will never observe this signal.

When the second pulse is delayed to the second sampling cycle, the oscilloscope will be triggered at the first rising edge or the second rising edge. There will be two waveforms on the oscilloscope screen, and both are at the trigger position. By adjusting the pulse delay time until the two signals can be triggered at the same time, the time taken for a complete sampling cycle can be estimated, and thus the refresh rate can be estimated.

3. Test steps

Based on the above analysis, a signal with a sufficiently long period and a particularly short pulse width is required as the test signal. Here, a manual Burst signal can be used as the test signal, which can basically meet the test requirements. Since the second channel of DG1022 does not support the Burst function, Tek's AFG3102 is selected as the signal source.

Set channel one to single-shot Burst. Since the pulse delay is limited by the signal period, set the signal period to 5ms (which means the default test machine refresh rate exceeds 200wfms/s), the amplitude to 1Vpp, and the pulse width to 5us. Also set channel two to the pulse word Burst, with a period of 5ms, an amplitude of 1Vpp, and a pulse width of 15us for easy observation. Set the oscilloscope mode to "Normal". Connect the two signals of the signal source to one input of the oscilloscope at the same time. Manually trigger the signal source to adjust the pulse delay of the signal source until the oscilloscope can trigger two edges at the same time, as shown in the figure below. At this time, record the pulse delay, and its reciprocal is the approximate waveform refresh rate of the current gear.