Inconsistent Resonant Frequency Readings w/ NI-USB-9234

[KU-Eng]

I am using a NI-USB-9234 signal acquisition device to obtain the resonant frequency of concrete specimens using an impact hammer and accelerometer.  I am getting inconsistent data after running my test numerous times.  I will get a different peak amplitude on my frequency response spectrum for each test.  I have ran this test using an older chassis and card owned by another department with good results.  The only differences between the favorable and inconsistent tests are that RSE grounding is used on the favorable tests and psuedodifferential grounding is used on the inconsistent tests.  Pseudodifferential grounding is the only option that my USB-9234 allows me to use.  Is it possible that this different grounding setting is causing my inconsistent readings?  I wouldn't think this would cause a problem since the frequency at the peak amplitude wouldn't change with different grounding, just the peak amplitude value would change.  If the grounding is the problem, is there a way to be able to use RSE grounding with the USB-9234 device?  Are there any other things that I could try to get repeatable data?

[Andrew.M]

Hi KU-Eng,

What are you comparing the data to that gives you the impression that your NI-USB-9234 is taking inconsistent readings? Do you have another device that is getting a more accurate signal? What is your working environment like? Are you in an area that has a lot of noise? What kind of lights are being used in the working environment?

Regards,

[KU-Eng]

Andrew,

The other equipment I have used in the past will give a consistent frequencies within 5-10 Hz as I complete multiple readings on the same specimen.  Also, the frequency spectrum is very smooth with no noise (smaller peaks) at any other frequencies.  My USB-9234 readings will give me frequency peaks at different frequencies as I complete numerous tests one after another on the same specimen.  For example, one reading might give me a peak amplitude at 1700 Hz while the next reading will peak at 2300 Hz.  Also, the spectrum has considerable noise (smaller peaks) throughout.  Another researcher felt that the extra noise found when using the USB-9234 is from using psuedodifferential grounding instead of RSE grounding.

My testing environment has been set up specifically to have very little outside vibrations.  The specimens are set on two single points on a rubber & foam apparatus to increase damping of outside vibrations.  The room in which the tests are being completed has typical fluorescent lights that you would see in an office/classroom.  I have heard that fluorescent lighting can cause interference at frequencies around 60 Hz, but my concrete specimens will have values between 3000-1000 Hz.

I appreciate the reply.

[Andrew.M]

Hi KU-Eng,

What other equipement have you used in the past that has given a more consistent frequency reading? The 9234 utilizes both analog and digital filtering, which is much different than most other devices. More information about filtering on the USB-9234 can be found here. What are you using to view the frequency spectrum? Are you using IEPE excitation on your device? Also, how do you know what frequency to expect? Are you running a controlled impact hammer? Are you using a coupling mode?

Regards,