strange fluctuations in voltage readings

[Tim3000]

Hi,

I am using a PCI6025E DAQ card to read in some voltages that I am trying to plot. When I use a volt meter to just manually see the voltages I am sending in the DAQ card they are very stable at approximately 5 volts. When I read them into the DAQ card and plot them I get sharp fluctuations of approx 0.5 to 1 volt around the 5 volt bias. I am not sure why this is happening. Any idea why this would happen or what I can do to stop it?
I have tried using the sample compression VI in order to take a mean of the signal to hopefully get a result close to what I am looking for, however I cannot seem to get that VI working for me.

Any help would be great. I can attach the VI if that would help.

Thanks,

Tim

[johnsold]

Many things can cause such behavior. External noise is the most obvious. The voltmeter readings make that less likely, but do not rule it out completely. High frequency signals averaged out by the voltmeter might be aliased into the data. Other things to look for are unterminated inputs, high impedance sources, and some inputs with much different voltages than others. Sampling rates, channel switching rates, gain switches between channels (if available on that board) can affect performance.

Tell us a few more details of both your hardware set up and post a copy of your program (perhaps simplified) that shows the problem. Someone may then be able to offer more specific suggestions.

Lynn

[shoneill]

Hi Tim3000,

I had a similar problem before which no-one was able to help me with. I was measuring a single channel (3 of 😎 on a PCI 6024 DAQ card (I can't remember the driver version, it was at my old job).

In the end (out of pure desperation) I tried reading out all channels, and extracting the value from the array of data. Much to my surprise this worked. I've as yet been unable to get a satisfactory reason for this, but the bottom line is that it worked for me.

So if you're reading a single, non-zero channel and are getting spikes, try reading out all the channels and extracting the corresponding values by software.

Try it out, it doesn't take much time, and you never know, it might just work.

Either way, I hope this helps

Shane.

[Tim3000]

Shoneill,

Thanks for that tip. I am not too great with LabView..yet. So to do this, I think I can obviously read in all of the channels (the same way I'm reading in one).However, to then extract the data from the channel I want how would i do that? Sorry that might seem really basic, but I am quite new to this.

Thanks

[tbob]

Use a scope instead of a DMM to see if fluctuations are actually there.

[shoneill]

Hi tim,

It's quite easy. You acquire all channels and then the output is an array of waveforms instead of a single one. You then use the Index Array function to extract the one you want, at which point you have a single waveform again.....

Hope this helps

Shane.

[DCSmith]

Are you making differential measurements? If so, do you have the 100k ohm resistors on the analog input channels of the daq card? (put 100k ohm resistor from each side of the input to analog ground)

[Tim3000]

Hi,

I tried reading all of the channels and then extracting the one with my data on it. It does not solve the fluctuation problem for me. Thanks for the tip though.

DCSmith, Could you explain what to do with the resistors a little more? I am not sure what you mean "from each side of the input". In the meantime I will try a few things.

Thanks,

Tim

[Caroline Bright]

Hello Tim,

Lynn is correct that there could be many reasons for this fluctuation. This fluctuation may actually be in the original signal and the voltmeter is averaging the signal which makes it appear that the signal is stable at 5V.
I would also like to know if you have used Measurement and Automation Explorer that comes with the DAQ driver to view the data in a test panel? This will allow us to determine if the fluctuation is in the original signal or is appearing because of the program. You can find help in setting up the device by opening Measurement and Automation Explorer in Program Files -> National Instruments -> Measurement and Automation Explorer. Once the program launches, go to Help -> Help Topics -> NI-DAQmx -> DAQ Quick Start Guide. Here you can find information on how to configure the device and then view the data in a test panel. You can then import these channels into your LabVIEW program without having to reconfigure the device.
Finally, are using one of the shipping examples that comes with LabVIEW or have you created your own program? If you would like to attach a screenshot of your program and a detailed description of your setup, I can try and reproduce the behavior you are seeing.

Thanks,
Caroline
National Instruments

[Tim3000]

Caroline,

I have tried using the Measurement and Automation Explorer and under Devices and Interfaces I have two DAQ devices. The Traditional NI-DAQ Device and just NI-DAQmx Device. When I look at the test panels of both of them I get the correct response when using the NI-DAQmx Device in Analog Input RSE mode. The other modes within the NI-DAQmx as well as when I use the Traditional DAQ are not correct. So, you mentioned I can input the channel from the Test Panel screen? How should I go about reproducing the correct response within my program. I will attach the program for you too take a look at if you like. To quickly explain my program, I have two for loops, the top one just sends pulses to a step motor I have which moves an antenna stand it also performs a channel read every few pulses. The data it reads is then plotted against the data from the other for loop, which is just degrees from -90 to +90. It is for a radiation pattern. Hope that helps.

Thanks very much

Tim