Overall Vibration

[RER]

Dear Patrick,

From the looks of you phemoninal forum title (Proven Active Veteran) you are very familiar to these forums, and a valuable member of the comminity.

I stumbled across your post, and would like to offer you a piece of advice. I would recommend that you start a new thread for this question. That way, if the members of the online community cant help you, it should then me answered by an applications engineer at national instruments. Reposting on an old forum prevents this happening.

Copy and paste your post into a new thread, and include a link to this particular thread if necessary.

Thank you very much for your time,

[pallen]

Thanks for the reply Rich.

After I made the reply in this thread, I found an interesting webcast here: http://zone.ni.com/wv/app/doc/p/id/wv-102

Great info there! 

Overall machine vibration is discussed right at the eleven minute mark.

But because of the rather immediate nature of the problem, I figured I'd try out the "Click for assistance" button on the page.  Sure enough my phone was ringing  and in a few minutes I was speaking with a DSA specialist in Austin.  I sent him some of my code via email, and he's going to check with some of his colleagues to either confirm that I'm on the right track or get me pointed in the right direction.

 

[dsb@NI]

Hi Patrick,

Whether you choose to use time-domain or frequency-domain integration is entirely up to your application requirements. In SVT, the time-domain integration is implemented via two filters (one to reject the DC component of the signal and one to actually apply the time-domain integration. Frequency-domain integration is a direct scaling of the input spectrum. The equation given by your colleague: ips = g*386.04/(2*pi*f) describes this scaling for the specific case where you are converting acceleration (g) to acceleration (in/s^2) and then performing single integration of the spectrum. You can probably get a better intuitive understanding of both time and frequency implementations if you have a chance to review the Integration book in the SVT user/concepts manual.

If you don't need the integrated waveform, I recommend that you use the frequency-domain integration VI because you don't have to wait for any filter settling, and it is trivial to band limit the result. You should still be aware that the default output of the frequency analysis VIs is in EU rms. You can use the SVFA Unit Conversion VI as shown in previous posts in this thread to convert to peak units. In LabVIEW there is no real limitation on the number of lines that you can have in the spectrum. Just wire the acquired vibration signal to the SVFA Power Spectrum VI. The longer the period of the acquisition block, the finer the frequency resolution will be in the measured spectrum. Finer frequency resolution may be desirable if you have identifiable tonal components in the vibration signal.

As for logging these results, you can use any number of file formats or write these results directly to an Excel spreadsheet. We can provide more direct guidance with a better understanding of your requirements.

Hope this helps.

[CUA]

Hi,

where can be found webcast you mentioned. Link seems to be broken.

Thank you.

[johnsold]

That thread is almost four years old so it is not too surprising that the link may have broken.

I suggest that you start a new thread and ask specific questions there about any problems you may be having related to vibration.

Lynn

[GUAU]

Hi Dough,

Would it please be possible to send the code for the overall vibration to me?

Thanks

[dsb@NI]

Guau, this is an old thread. If you need help acquiring vibration signals and/or calculating overall vibration level, please start a new post. For the community to help you, please include as much information as you can. For instance, how many sensors? what type? acquisition device? hardware configuration? analysis requirements? 

All of the screenshots in this thread include code that ships with the LabVIEW Sound and Vibration Toolkit. Install it, evaluate as necessary, and let us know (by creating a new post) what questions you still have.