Sound Power Measurement with 4472

[Gurdas]

Is onboard DSP a must for any Sound Power Measurement?
I know that NI-Sound Power Plus works only with 455x.
Is it simply because NI has not yet written a software for Sound Power Measurement using 4472 or is it technically NOT possibly to do Sound Power Measurment using 4472?

[AVR]

Onboard DSP is NOT a must for sound power measurement. You can acquire a raw sound signal with any high-quality DAQ card combined with proper signal conditioning. Filtering and correlation (simultaneous sample-and-hold) requirements may influence your final hardware setup. Ultimately, however, the entire operation would depend upon the software for processing and analyzing the raw voltage data. One benefit of this method it that it allows you to archive the unprocessed data. What the onboard DSPs and anti-aliasing filters provide (at significantly more cost) is a less-numerically intensive, and must faster, means to acquire the desired data. Many organizations do not have the support staff to develop and maintain the software required to perform the analysis; s
o, a better solution is to purchase tha hardware that will perform most of the analysis operations "on the fly".

[Gurdas]

Thanks for the answer Mike.

We are lookng at developing solutions to address the NVH market. For platform, we zeroed on:
1. Dolch (www.dolch.com) NotePAC Plus for NI-4472 and FieldPAc for 455x or 445x
2. Magma (www.magma.com) with any Laptop for NI-DSA or NI-DAQ as per requirement

However, we would also like to develop a very affordable single channel solution for say just sound pressure OR sound power OR vibration measurements. Here, we'd like to use Multipurpose DAQ. Can you suggest which NI DAQ fits the bill best keeping in mind the application and cost? I feel the newly launched 16-bit 6036 is right. Would love to have your insights.
Best regards,
Gurdas

[Giarc]

AVR is certainly correct, you do not need onboard DSP to do sound power measurements. With the higher speeds of today's processors, many applications don't require onboard DSP even if you want to perform processing on the fly (which traditionally has been done with DSPs because it is quite processor intensive). NI actually packages the 4472 with the Sound & Vibration Toolset, which includes VIs for performing sound power measurements, plus many other VIs for performing other sound and vibration type measurements and analysis. Check out the link at http://sine.ni.com/apps/we/nioc.vp?cid=5031〈=US that lists the pricing of the 4472/SVT package.

You can also get additional information on the toolset itself at http://sine.ni.com/apps/we/nioc.vp?cid=312
2〈=US .

Just a quick note, the 4472 does have antialiasing filters built into it to help improve signal quality.

Hope this helps!
criag

[Gurdas]

Thanks Criag. If you are well into NVH I would love to have you mail me at

As I mentioned in my comment to Mike's (AVR) answer, we are also interested in developing Laptop based low cost NVH system. One suggested config is NI-6036 DAQCard with ICP signal conditioning (CA-1000 + SC-2345 + SCC-ICP01).

I'd love to have two inputs from you here:

1. What are the major issues for the above mentioned low cost system? Say I keep it to just 1 or 2 channels and restrict to Sound Pressure and Vibration measurements with FFT. For 2 channels, a skew of a few microsecs is fine with me.

2. NI has a software called Sound Power Plus (very expensive). I was under the impression that ONLY this gives Sound Power and that the Sound and
Vibration Toolset has NO tools for sound power. Am I wrong?

Best regards,
Gurdas

[AVR]

1. The 6036E is a "sweet" card for high-speed, high-accuracy DAQ on the PCMCIA bus. Even at 10-times oversampling you can achieve 15-20 kHz of response with nice line resolution in the FFT. My experience with the PCMCIA bus has shown a wild variation in DAQ bandwidth (IBM/Compaq...the PITS -- Dell...the best ; and , no I do not work for Dell). For signal conditioning, I use SCXI (so I can not comment on the use of the other components). If you can live with interchannel skew, the set up looks quite promising. Our DAQ requirements demand the use of simultaneous sample-and-hold (SCXI 1140) for either two-channel hot-film measurements or multiple sound pressure measurements with precise cross-correlation (to eliminate wave reflections and e
chos within the test article). Our microphones were developed by NASA for use in VERY HIGH temperature environments, so our signal conditioning ahead of the 1140 is made from custom devices.

2. It is important to note that any toolkit add-on to LabVIEW "could" be written by the end user using the very same subVIs distributed with the Full Development System. Sound power is merely a term used to describe a particular mathematical method involved in transforming the acquired sound data from the real domain to the frequency domain. The mathematical methods have been used for many decades. I use my own custom code originally developed in FORTRAN, converted to C++, and now embedded in a code interface node (CIN). The issue to resolve comes down to a) do you have the in-house capability to write and support detailed code development, or, b) do you purchase the toolkits and thereby achieve the capability very quickly. Although the toolkits are expensive, the burdened cost of, say, 1
additional highly-skilled employee could cost 20 times as much.