Specifications

Sampling

Because the idea of sampling is so fundamental to the operation of a digital oscilloscope, let's look at it

in more detail.

An analogy for sampling is to think of the A/D converter as a camera that takes a picture of the

amplitude of the waveform. The "shutter time" is very short and the camera "assumes" that the

waveform is constant over the time that the "shutter" is open.

As an example, you might behave like this "sampling camera" if you were monitoring the outside air

temperature with a thermometer. You'd look at your watch, write down the time, then write down the

thermometer's reading.

If your thermometer was near the exhaust of a heat engine, the air temperature could vary more

rapidly than your thermometer is capable of responding. You would reconstruct the temperature

waveform as a function of time by using your samples -- and, depending on your knowledge of what

you were measuring, you may or may not be aware that your reconstruction of the waveform might

not accurately represent the "real" air temperature. The term "real" has quotes around it because we

can only know it through measurement -- and different measurement methods have different

capabilities, leading to different "versions" of reality.

Let's look at two cases of sampling, one where the sampling is adequate and one where it is not. These

qualitative examples will give you a feel for what good sampling can be like. Here's a waveform that

was constructed by a program that used 1000 points to display the waveform. The "Adequate

sampling" waveform shows the sampling points taken by the "camera" at 1 out of every 20 points in the

waveform. The "Inadequate sampling" waveform was 1 out of every 10 points.

Figure 19

Copying/reprinting/republishing without express written consent prohibited.