power line noise of NI-9205

[Sebastian.Peng]

We are trying to measure the output of a highly sensitive sensor with the NI-9205 converter and MATLAB as interface. The signal voltage is typically a few mV ,with a theoretical noise voltage of a few uV). But the signal we acquired is interfered by a much larger noise ,which can be clearly identified as a time-varying power line noise with an amplitude of 20mV, as shown in the time domain plot and the power spectral density plot, which can also be observed in the short-circuit channel. The noise could disappear occasionally.

The DAQ configurations can be seen in the pic: the range is set as ±0.2V, the terminal config in differential mode and the coupling mode is DC.

To find out the potential causes, we used a in-house constructed DAQ converter as a contrast and surprisingly the signal was pure with little noise.The following pics show the diagram of the system and the environment of the working site.

[ZYOng · Accepted solution]

I guess that your sensor is a floating signal source. When using a differential connection with a floating signal source, you need to connect bias resistors. See Table 1 in Field Wiring and Noise Considerations for Analog Signals

You can try to use Referenced Single Ended (RSE) mode to see if the noise goes away.