Signal Conditioning

cancel
Showing results for 
Search instead for 
Did you mean: 

Noise Problem in Microphone Signal from SC-2345/SCC-ACC01/DaqCARD-6036E

Hello, I am trying to acquire sound pressure level data using an SC-2345.  The setup is:
2 NTI miniSPL microphones into SC-2345 using 2 SCC-ACC01 Modules then into a DaqCARD-6036E.

Details:
SC-2345
J1 = SCC-SG04
J2 = SCC-SG04
J3 = SCC-TC02
J4 = SCC-ACC01 (mic 1)
J5 = SCC-ACC01 (mic 2)
J6 = SCC-AI06
J7 = SCC-AI03
J8 = SCC-AI03
J21 = SCC-PWR03 run on 12V wall-wart in the lab and 12V car battery in the field.

The inputs are set as NRSE and grounding is to the grounding lug in the SC-2345.  The Daq E series user manual says to connect the -'ve of NRSE inputs to AI Sense (page 2-21), but with only one AI Sense input on the SC-2345, I get lots of cross-talk with both -'ves plugged in there.  Also AI Sense and chassis ground are tied together.  As a result, I used the alternative setup and have +'ve to terminal 2 and -'ve to terminal 1 on the SCC-ACC01 (page 2-24 of Daq manual).  Here is an image that should help to explain the arrangement.

The noise that I am getting is broad spectrum, but there are some spikes.  Here is an image of an a-weighted (human ear response weighting) FFT of the data collected in a quiet room.  Notice the noise spikes at 4000Hz, 6500Hz, 7200Hz, 7900Hz and 8800Hz, these tend to wander around while the unit is warming up.  A quiet room should be about 50dB and I'm getting 10dB higher in just bothersome noise.  This comes through in playback as a high pitched whine that is nearly as strong as the signal of interest.

Does anybody know anything that can help me?  If you have any additional questions, please let me know.

Cheers,
Seth


0 Kudos
Message 1 of 4
(5,284 Views)
Hi Seth,

I would suggest following the ACC01 manual to connect your microphone. I also posted the connection diagram, see below. The DAQ manual is showing connections for regular breakout boxes, unlike the ACC01 which is setup in Measurement and Automation Explorer (MAX) and has its own connection diagram.


Message Edited by David L. on 07-13-2007 05:15 PM

David L.
Systems Engineering
National Instruments
0 Kudos
Message 2 of 4
(5,269 Views)

I have the same problem, only I am already using the connection shown in the SCC-ACC01 manual, and suggested above by "David L."

In my case, there are stationary frequency components around 2, 3, 10 and 15 kHz, and playing back the acquired signal, I too noticed a high pitched whine as strong as the signal of interest.

Any ideas? Thanks!

Rodrigo

0 Kudos
Message 3 of 4
(4,891 Views)
Hello Rodrigo,

The first step is to figure out where the noise is being introduced.  Is the noise present when there is no input waveform (i.e. no sound being input)?  What is the magnitude of the noise?  When you say the noise is stationary, does this mean its magnitude and frequency do not change with your test signal?  Or possibly that the noise does not move over time? How long are the leads from your microphone to your 2345? Are these shielded?

Depending on the signal you are measuring, depending on how you are programming, you could implement a software filter.  LabVIEW and SignalExpress have built in filter vi's you can specify a cutoff frequency below the noise.

Once I we figure out where the noise is coming from, it will be much easier to troubleshoot.  I look forward to your response.
Neal M.
Applications Engineering       National Instruments        www.ni.com/support
0 Kudos
Message 4 of 4
(4,880 Views)