analog input phase drift

[jmountney]

The phase of analog input signals are drifting over time in my system. A 1kHz square wave with 50% duty cycle is sampled at 50kHz with the 9205 module and cDAQ-9178 chasis using the Voltage-Continuous Input example VI. While observing one period of the signal in the waveform graph,  it can be seen that the square wave is slowly shifting from left to right in the display. An example of this phenomenon is given below. Changing the clock source from "on-board" to "PFI0" casues the drifting to slow down, but it is still present. The input signal was measured with an oscilloscope and found to be time-stable. The input source, DAQ, PC and external clock all share a common ground.

What would cause the sampled signal to shift over time?

Are there software buffers that need to be reset?

Are there any known LV time stamp misallignments?

Initial waveform

Waveform after 10 s of sampling

[Henrik_Volkers]

How do you sync the 1kHz generator with your DAQ?

What you see is the drift of two free running oscillators. One from your DAQ and the other from your genrator 🙂

The scope show you a stable waveform if you use the trigger. If you choose a free running mode and don't use the calibration squarewave of the same scope which migth share the same clock, you can and will see the same effect.

What do you want to see? Just a stable square? -> Trigger your DAQ