Issue 140 March 2002
Spy-Size Event Logger

Periodically

Periodic sampling must be used whenever the data does not contain event (logic state) information. Analog data doesn’t contain event information, unless you’re interested in zero crossings or some other such signal conditioned output. Data samples must be taken at the Nyquist rate of no less than twice the highest frequency of the data. The highest data frequency is that which you are interested in (providing all higher frequency data has been properly filtered out prior to sampling). This is why it is extremely important that you know what it is you’re interested in. Although the data logging of sensor outputs is generally considered low frequency (below audio), you can certainly see how this is applicable to higher frequencies.

Because a periodic data logger samples and stores data at defined intervals, you can calculate the time of each data point referenced to the start of data logging by multiplying the sample number by the sample rate, so only data needs to be saved. However, because of the periodic nature of this mode, logging memory is consumed at a steady rate. The upside to this is that only the actual data or change in data needs to be stored, no time stamp is necessary.