Capture time, the usable signal duration shown on an oscilloscope screen, is related to the scope's maximum sampling rate and the length of the acquisition memory. LeCroy's WaveSurfer oscilloscope has a 250 kpoint per channel acquisition memory which, in two channel mode, provides 100 times the capture time at its maximum sampling rate compared to a similar scope which only offers a 10 kpoint memory. For example, WaveSurfer provides up to 200 µs capture at its maximum sampling rate of 2 Gigasamples per second(GS/s) compared with only a 2 µs capture time for a scope with only a 10 kpoint acquisition memory.

Lets look at the real consequence of this. Consider the serial peripheral interface (SPI) signal shown in figure 1. The waveform is captured using a LeCroy WaveSurfer scope with the standard 250 kpoint per channel maximum acquisition memory. The memory channels are interleaved to provide dual channel with 500 kpoint memories. This display uses a capture time of 1 ms to see two adjacent data bursts on trace C2. Trace C3 shows the corresponding clock signal. Expanded (zoom) traces of both waveforms are shown in the lower traces marked Z2 and Z3, respectively. The resulting sample rate of 500 MS/s provides excellent time resolution and sufficient sample rate to resolve the signal. Now consider Figure 2. This shows the same signal acquired using a scope with only a 10 kpoint memory. For the same capture time the sampling rate is only 10 MS/s. Note the loss of detail in the clock pulses and the poorly defined transition times of the data signal. This is caused by the low sample rate which effectively limits time resolution of the measurement. Although the unzoomed view of both scopes look similar, it is obvious that the LeCroy WaveSurfer provides more information about this signal.

Figure 1:

A measurement of a 1 ms capture time sampled at 500MS/s using 500 kpoints of memory. The high sample rate produces excellent fidelity.

Figure 2:

The poor performance of a scope with only 10 kpoint of acquisition memory due to a lower sampling rate of only 10 MS/s

From this example you can see why capture time is important. You will need long capture time for several reasons:

  1. If you require the ability to capture a mix of signals that are spaced far apart in time
  2. You need to view a long pre-trigger time interval.
  3. You need the ability to capture both high frequency and low frequency signals in a single measurement
  4. There is a long delay between the trigger event and the measurement point.

To effectively use a scope over a wide range of capture times it needs to have adequate acquisition memory to do the job. Figure 3 shows the relationship between capture time and sampling rate for the WaveSurfr and another scope in its class with 10kpoints per channel memory. For a capture time of 1 ms, marked by a dotted red line, the WaveSurfer has a sampling rate that is 50 times higher MS/s vs.10 Ms/s. This means that the scope with the longer memory has a time resolution 50 times higher than the scope with the shorter memory.

Figure 3:

A chart showing sampling rate as a function of capture time

It's not enough for a scope to have a high maximum sampling rate. Without a long enough memory the capture time at the high sample rate will be limited to very short captures.