Wrong measurement with a PCI-4472 for vibration analisys

[Luis Medina]

I´m using a PCI-4472 with an Krüel & Jaer Triaxial Piezoelectric accelerometer (ICP excitation anabled), and Labview 6.1 with Sound & Vibration Toolset.

Comparing the measurement with a Krüel & Jaer vibration system (using a same sensor) the amplitudes of the PCI-4472 are different and the frecuency is ok... the funny thing is that changing the sensivity values (from 10mV/EU to 10000 mV/EU) in the Labview Example Program, some times the amplitudes became similar to the Krüel & Jaer analysis system, if I want to read good signasl I have to keep the computer turn on forever.

I´m using the same parameters for the FFT analisys, ICP excitation. I´m working with a Celeron Processor with 128Mbytes RAM with Windows
98. The firmware of the DSA board is actual...

Carlos Rios (carlos.rios@ni.com) our support engineer came to evaluate the problem, and we need to solve the problem in this week... please help..
Luis Eduardo Medina

[Gerald Albertini]

Luis,

The sensitivity value you set in the sound and Vibration Toolset will allow your signal to be displayed in an engineering unit.
An acquisition board like NI-4472 will return "Volts" when you're interested in "g" and so, you have to convert from Volts to g.
This is done for you in the Sound and Vibration Toolset. In order to make the sensitivity of your accelerometer the default value, simply use the "Operate --> Make Current Values Default" command in the LabVIEW menu bar.
I hope this will help.
Gerald

[Luis Medina]

Hi Gerald, thanks for your comment. As you comment we are using the "SVT Voltage to EU" VI, to transform the measure units to m/s^2 for example. Our sensor has a sensitivity of 10mV/m^2 and 100mV/g. This values are default in the application.

And the PCI don´t work. We are trying with another PCI-4472 to anulate the posibility of hardware failure.

We are focusing in the sensor, the data sheet especifies that all the three channels have to be powered for measuring.

[Gerald Albertini]

Luis,

Did you make sure AC Coupling was turned on ?
When ICP excitation is turned on, AC-coupling needs to be enabled.
Did you try to run a Sound and Vibration example on a single channel to test your system ?
I would recommend you try to use the example called "SVXMPL_Baseband FFT (DAQ).vi" located in the "Program files\National Instruments\LabVIEW 6.1\examples\Sound and Vibration\Frequency Analysis" directory.
This example will display a popup message that will ask you if you want to turn ICP excitation ON and will set AC Coupling accordingly.
It will then display the Power spectrum measured on the selected channel.
I hope this may help you troubleshoot this problem

Regards

Gerald

[Luis Medina]

Hi Gerald,

I´m using the SVXMLP_Vibration Analisys (DAQ).VI to test the sensor and the DSA.

And I´m following the steps you said: AC coupling and ICP enabled.

We have the SMB to BNC connector attached to an adapter from Brüel & Kjaer.

We´re working on the sensor config and cable to ensure the connections are compatible.

I´m sending you the data sheet of the sensor (DeltaTron Accelerometer type 4506).

We have an ID report of the PCI-4472 failure: 458617
With Gerardo Garcia from national instruments. But your sugestions are welcome.

Thanks a lot!

Luis E. Medina

[Gerald Albertini]

Luis,

The first thing to do is determine if the board is out of order.
In order to do so, you should :
1./ Check the version of NIDAQ you are using
2./ Launch the test panel in Measurement and Automation Explorer (MAX) and make sure the board is working properly (ICP is not supported in MAX)

You're mentioning an ID report of the PCI-4472 (458617)...Are you talking of the error code returned in LabVIEW ?

Gerald

[Luis Medina]

Gerald,

We tested the PCI-4472 with a function generator and the software (MAX & Labview) measures the voltage signals well. So the problem could be in the sensor.

We´re going to connect another piezoelectric accelerometer in order to locate the problems.

The ICP excitation works for all the channels?

The ID report involves the problem with the PCI-4472.

[Gerald Albertini]

Luis,

I've been looking at the code of this example and the excitation will be enabled on the channel that has been selected (default value = channel 0)
ICP excitation is available on a "per channel" basis

ICP excitation is enabled (or disabled)using the AI Parameter.vi.
The channel string may contain multiple channels (ex : "0:3" to enable ICP on channels 0 to 3)

It should be very easy to modify any example from the Sound and Vibration Toolset to accomodate a triaxial accelerometer measurement.

Please let me know if you need assistance with it.

Gerald

[Luis Medina]

I´m using an amperimeter to verify the ICP excitation.
I´m using "Level Meter (DAQ).VI" to activate the ICP on each channel.
The problen is that I need to run the VI two times to measure current across each channel.

I don´t have the ICP excitation enabled on the three channels using the AI Parameter.VI as you said. I need to activate the ICP "manually" for each channel...

And good news, the system record good measures when the ICP is enabled on the three channels.

I´m going to need assistance with the VI.

Thanks

Luis

[Luis Medina]

Gerald

We programmed a routine in Labview to force the ICP to be enabled on each channel (0,1,2) (with the Get DAQ info.VI). I can´t explain the need to execute the ICP two times to observe the current outputs in the channels.

Meanwhile we are working on the program app.

Thanks for you support
Luis E. Medina