NI 9233 Microphone Signal Conditioning

[Midlothian]

Following an update to LabVIEW 2009 the Pre-polarised GRAS microphones no longer work correctly with the NI 9233 multifunction DAQ modules. (They have been used regularly since 2007).

There is a significant distortion of the calibrated signal showing clipping of the positive part of the trace at higher input levels.

Connecting an external signal conditioning system to the microphone shows perfect sine wave using the same microphone capsule (GRAS 40AE), amplifier (GRAS 26CF), cable, and NI 9233 - we have 2 examples of each item.

The only item that seems to be common to the fault is the signal conditioning from the NI 9233 following the LabVIEW update from 8.6 to 2009. I have tried changing the level of excitation from 2mA to 4 then 8 without any benefit (the specification from GRAS says 2-20mA nominal 4mA). I noticed that there are some changes in the terminal configuration option dialogue but having tried to change this there is only one viable option "Pseudodifferential".

The attachment of external power sources and excitation for the microphone is not an option for practical reasons but it does restore the original function.

Is there a documented change to the 9233 function with LabVIEW 2009 or is there a simple solution to this change in behaviour?

Thanks.

[Midlothian]

P.S.

Further investigation of the same NI 9233 MxDAQ shows that the conditioning appears to work with an accelerometer requiring internal excitation.

Why would the microphones with the same requirement for 4mA no longer operate without an external excitation source?

[sahpe]

Hello Midlothian,

As you have also emailed our Technical Support team, let's work on this through that channel and once we have it figured out we can update the community of our findings.

If any members have input on the subject, please feel free to post anyway.

Kind Regards,

Michael S.
Applications Engineer
NI UK & Ireland 

[Midlothian]

Hello Michael,

Thanks for the reply.

I too hope that we can resolve this through the support email channel.

For other members input I have the following update: 

I have now confirmed that the output of the NI9233 is restricted to 2mA and also verified with independent instrumentation that 2mA (DC) is passing by both the NI9233 and when the external excitation is introduced. The addition of the external excitation completely clears the clipping problem and the calibration shows that the microphones are both within their correct sensitivity range.

[Midlothian]

I have rebuilt an old 2007 test laptop with 8.2.1 NI software and shown that the decay problem and the apparent signal clipping of the positive wave only are (were) apparent if the mic. calibration is run for more than a about 10 seconds in the calibrator.

The signal becomes clean again if:

1. the microphone is removed from the calibrator and reinserted,

2. the cable to the microphone is interrupted and rejoined,

3. the USB is disconnected and the vi is restarted.

4. the external source of excitation current at 2mA is added.

This may be a hardware-related but it does involve two NI9233 and two mics.

It is possible that the problem escaped attention previously since calibration only takes a short time to perform.

[Midlothian]

Further checks:

"Bias" Voltage across the NI9233 is measuring 15V when operating with 114dB sound source.

120dB = 20 Pa (ref. 20 microPascals) so 114dB is 10 Pa.

With a sensitivity of 50mV/Pa that is 500mV of signal over the 15 V.

The maximum compliance voltage of the NI9233 would need to be below 15.5 V to see this clipping effect?

[sahpe · Accepted solution]

For the interest of other users, it seems that the microphone needs slightly more excitation at 114dB, hence clipping when used with the 9233 that can only provide 2mA. The setup works fine at 94dB, where 2mA is enough excitation. There are no errors appearing in software as this is a hardware limitation.

Kind Regards,

Michael S.
Applications Engineer
NI UK & Ireland

[Midlothian]

I have confirmed that the NI9233 supplies the 2 mA by measurement of the current in the cable and yet shows clipping at 114 dB. 

I have also confirmed with 2 mA provided from an external source, that there is no clipping of the signal. It is a hardware issue with either the NI 9233 or with the microphone.

Note:

The VI software selectable options for controlling the NI9233 IEPE are not capable of influencing the behaviour of the hardware. This is covered in the pdf operating instructions:-  

The NI 9233 IEPE excitation current and AC coupling are always enabled.

IEPE excitation current

Minimum....................................2.0 mA

Typical ........................................2.2 mA