usb-6009 acquisition problem

[marcoalba]

Hi

I have a problem with the acquisition of two signals coming from two sensors. The signals come from an accelerometer and from the impulse of a hammer.

The signals pass from the sensors to a conditioner and then to the DAQ.

I put RSE mode when I create the virtual channel.

Both signals have the same noise that does not appear if I measure the signals by oscilloscope before they get in the conditioner.

The time-history of noise of both signals has this shape,

Then when I give an impulse with hammer, the accelerometer can overcome this problem because the peak is around 6Volts. Instead, for the force signal this noise is not neglectable, compared to the peak(around 0.3 V), and the FFT are completely wrong.

The offset is not a problem for me at the moment but this noise doesn't permit me to get acceptable results.

Can you help to solve this?

Thank you.

Regards,

Marco A.

[PackersFan]

Whatever is causing the noise appears to be either part of the environment or your circuit (as both signals display the same noise).

Why are you using RSE instead of differential?  Differential signals are great at canceling out noise that appears on both lines.

[marcoalba]

I tried to change from RSE to differential and offset dissapears.

The noise remains.

Maye it can be helpful if I put other pictures of the problem.

Here are the signals and their FFTs:

Here are some close up of the signals and a test panel on the channel 0:

As you can see the results from accelerometer can be accepted but that's not the same for the force channel where the noise is not neglectable compared to the voltage of the phenomen I'm studying.

If I try to connect an oscilloscope to the signal that gets out from conditioner I can observe a noise  that seems completely different (much higher frequency and much lower amplitude) from the one I observe on the Labview.

Is there a way to solve this?

Thank you a lot for your answers,

Marco

[johnsold]

Marco,

Some of what you are seeing is the limitation of the USB-6009.  On the 10 V range, single ended, the resolution is 2.4 mV.  That is what your images show.  The specifications also say that the system noise is 5 mV rms.  Your data is well below that.

Your acceleration signal is large enough that you cannot change ranges.  You might try a more sensitve range for the force signal (if the USB-6009 allows different channels to ahve different ranges). Otherwise you will need an external amplifier for the force channel.

The force signal looks pretty clean to me, especially if it is a hammer impluse.  What are you trying to measure that the noise is too large?

Lynn