From Friday, April 19th (11:00 PM CDT) through Saturday, April 20th (2:00 PM CDT), 2024, ni.com will undergo system upgrades that may result in temporary service interruption.
We appreciate your patience as we improve our online experience.
Turn on suggestions
Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.
Showing results for
Topic Options
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page
Is there a roll-off with the SVT octave analysis vi's?
Alv
Member
10-27-2004 05:57 PM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
I built a spectrum analyzer that displays 1/3 octave and narrowband data. I'm using the PXi chassis with 3 NI4472 daq cards. I noticed when putting in low frequency tonals (from function generator, amplified to 1 Vrms) the SVT 1/3 octave analysis vi has a roll-off associated between 10-100Hz. Narrowband (using power spectrum vi) quantifies the correct level extactly. I've noticed the SVT Octave analysis vi's roll-off the amplitude levels by as much as 50-60dB at 10 Hz. Does anyone know the solution to this problem.
dsb@NI
Active Participant
10-29-2004 09:21 AM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
Could you post a sample of the code that demonstrates the issue? Are you applying weighting (A-weighting would apply a relative response of -70 dB at 10 Hz) to the signal when you do octave analysis?
When I input a 10 Hz tone from a function generator to to AI0 of a 4472, I get consistent results between the power spectrum and 1/n octave (see attached results).
Another thought, how did you implement 1/3 octave analysis? The 1/3 octave analysis VIs take an enum for the low frequency where the lowest frequency in the enum is 20 Hz. I had to use the 1/n octave VI which takes a DBL for the start frequency.
Hopefully, some of this rambling helps,
Doug
When I input a 10 Hz tone from a function generator to to AI0 of a 4472, I get consistent results between the power spectrum and 1/n octave (see attached results).
Another thought, how did you implement 1/3 octave analysis? The 1/3 octave analysis VIs take an enum for the low frequency where the lowest frequency in the enum is 20 Hz. I had to use the 1/n octave VI which takes a DBL for the start frequency.
Hopefully, some of this rambling helps,
Doug
Doug
NI Sound and Vibration
NI Sound and Vibration
LoganS
Active Participant
10-29-2004 12:01 PM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
I'm copying below an example program that has been modified to satisfy this issue. The original, SVXMPL_Third-octave-Analysis (Traditional DAQ).VI uses the SVT VI for Third-Octave Analysis, and is limited to ranges from 20 Hz to 20kHz. The changes I made in the VI include replacing the Third-Octave Analysis VIs with Fractional Octave Analysis VI, setting it to be a Third Octave opteration, and changing up the inputs to it to allow you to choose any range you desire (within reason). Good luck, and have a great day!
Logan S.
Logan S.
10-29-2004 01:55 PM
Options
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report to a Moderator
Logan,
Thanks for giving me a call. I reviewed the code you recommended, however the code is much the same as mine. I still have concerns regarding the noise floor around the calibration signal I put in. When I use the VI you suggested, the noise floor looks normal (but thats one channel). I have the very same vis, DAQmx to scale VI then to SVT fractional Octave VI (for 1/3octave 10 channels). I also have a problem of the filters not settling, and maybe thats the cause. Ross is looking into that issue.
Thanks for giving me a call. I reviewed the code you recommended, however the code is much the same as mine. I still have concerns regarding the noise floor around the calibration signal I put in. When I use the VI you suggested, the noise floor looks normal (but thats one channel). I have the very same vis, DAQmx to scale VI then to SVT fractional Octave VI (for 1/3octave 10 channels). I also have a problem of the filters not settling, and maybe thats the cause. Ross is looking into that issue.
