Removing the analog signals changes my frequency measurement

[Thanos12]

I am trying to record signals from four sensors (analog output, differential, 0-10VDC) connected to the AI terminals of a NI-6215USB, simultaneously with the signal from an anemometer. The anemometer generates a pulse output and I connect its + cable to the PFI1 (Gate) terminal, and the - cable to the DGND terminal for a frequency measurement. When everything is connected I get correct readings for the analog channels and erroneous readings for the measured frequency. When I disconnect the analog signals completely from the terminals, I get correct frequency readings but a residual alternating voltage is also logged in the analog channels.

I don't know if it's matter of wiring or it is impossible to simultaneously acquire several analog and counter signals. I have attached two screenshots (from Signal Express 2015 that I am using). The first one (analogs in) shows the case where the analog signals are connected, but I get unphysical high noise-like signal in the frequency measurement, in the order of thousand Hz. The second one (analogs out) shows the case where I disconnect the terminals from the analog sensors, I get the correct expected frequency of around 83Hz coming from the anemometer (with the exception of some spikes) but there is also an alternating-like voltage in the analog channels (where no sensor is connected at all).

I have trying changing the sampling period in the Counter Frequency Step Setup panel, but nothing helps. I also tried connecting the + cable of the anemometer signal to the PFI0 (Source) terminal, but the result is still the same.

[Shalini.M]

Hi Thanos12, 

Do you have your AI GND wired?
How exactly do you have your four sensors wired in the AI terminals? 
Also, when you are measuring frequency, what exactly do you have wired? 

[Thanos12]

Hello Shalini,

thank you for your response. Each (+) cable from every one of the four analog sensors is connected to a separate AI terminal. Their (-) cables are all connected to AI GND. The have been set as RSE sensors and I get correct readings with this setup.

The problem is when the anemometer signal is connected to the counter terminal and i start the frequency measurement. The (+) cable from the anemometer is connected to PFI1 and the (-) cable to DGND. The measured frequency is then uphysically high, with lots of noise, and the only way to get a correct reading from the anemometer, is to disconnect the AI terminals.

If it's any help, I have noticed that only when the anemometer frequency is very low (around 10Hz and decreasing), then the measured frequency is correct. When the anemometer rotates at higher air speed, generating a signal of higher frequency, the readings are wrong. I am expecing signals in the range of 70-350 Hz from this particular anemometer type. I have tried various combinations of measurement method and time, one or two counters, etc, but I still cannot get the expected signal.

In the attached picture, you can see the signal from the anemometer that is somehow better below 50Hz, presenting some downward spikes, but still better than the totally chaotic signal in my previously posted pictures. Below 10Hz the signal is more believable (this is a capture of a signal from the anemometer that was rotated and left to come to a stop).

[Shalini.M]

I have a feeling the issue might be related to wiring. I would recommend taking a look at this document that explains field wiring and noise considerations for analog signals: http://www.ni.com/white-paper/3344/en/