10-24-2008 07:29 AM
We recently bought two boards PCI 4472B. One of them has a channel (#5) that is working different compared to the others and the other board also has a difference in the channel 3 but with lower magnitude.We made the following steps: 1) Both boards were installed in a computer with 2 PCI sockets. The software used is NI-DAQmx 8.0 and LabView 7.0 Professional (Sound and Vibration toolset 3.0)2) The boards were installed under recommendations3) The tests on each board were performed using 3 triaxial accelerometers (100mV/G’s) using AC Input Coupling and IEPE Signal Conditioning Enable (4mA)
Attached you will find the graphs obtained (amplitude in m/s2): Attachment01As it can be seen on the attachment the cannel #5 (Acceleration4) has a high negative offset.
Additionally was performed a self calibrating (with NI DAQmx) on the most problematic board and the results can be seen on the attachment02. At the other board, since it was not measuring correctly it was performed the self calibrating getting the results shown on the attachment03.The acquisition was set up as it can be seen on the attachment04. When we measure with no accelerometer connected the signal obtained can be seen on the attachment05. It can be seen that the channel #3 has an unacceptable offset and the other channels are better but not centered at cero. Then we performed some measurements using the following hardware:
10-24-2008 08:55 AM
10-28-2008 06:59 AM
If you don't mind, I'd like to have you call our support line at 1-866-275-6964 to resolve this issue. From the look of your graphs and the fact that the calibration failed on one board, it looks like we might be having issues with the card itself.
07-02-2012 05:06 AM
Dear marcos ，I have a question about PCI -4472B ，Can you help me？
We use PCI 4472B didn't connect any accelerometers using AC Coupling and IEPE signal Conditioning Enable(4mA),
We can see from the monitor that the amplitude is not 0mV but 12~19mV(every channel is not the same)，
then we connect the 3 triaxial accelerometers (10mV/g) to PCI 4472B with BNC to SMB，and hold the 3 triaxial accelerometers still，
we can see from the monitor that the amplitude still is not 0mV why ?
07-02-2012 12:05 PM
Are your 3 triaxial accelerometers externally powered? They likely are if you aren't using the IEPE excitation, but just making sure that something is powering the accelerometers.
Are you using the 4472 in AC or DC coupled mode?
What amplitude do you see with the accelerometers connected? You mention 12-19mV, but it's unclear whether these readings are with or without the accelerometers connected.
The maximum 4472 specified residual DC-coupled offset is 3mV.
If you have a 50Ω terminator, connect it between AI± and let us know what offset voltage you measure.
Note that the 4472 is ideally suited for sound and vibration applications, which require precision, not DC accuracy.
Let us know the details above and we can look into your situation.
07-02-2012 01:26 PM
We sent the boards to recalibrate when we experienced those problems but they returned with the same condition you mentioned (approximately 20mV offset). We solved the issue grounding the boards to the computer chassis and eliminating the mean value by software. Hope this helps and let me know if you have further questions. Thanks!
07-03-2012 03:48 AM
Hi John S：
We use the interior powered （4mA）， in AC coupled mode.
Without the accelerometers connected obtained the graphs: Attachment01;
We see that Yesterday the amplitude was 12-19mV，but today it's 9-13mV.
With the accelerometers connected to AI0(X)\AI1(Y)\AI2(Z) the graghs obtained:Attachment02
The amplitude decreased to 6-7mV which connected the accelerometers，the other decreased nearly 1mV.
Also we have one USB-4431，we connected it to the same computer and set it to AC coupled mode use interior powered （2.1mA the max current the USB-4472 can support）， then we obtained the graph as this：
Without the accelerometers connected the graphs obtained: Attachment03，
With the some accelerometers connected to AI0(X)\AI1(Y)\AI2(Z) the graghs obtained:Attachment04.
07-03-2012 06:45 PM
First, I notice that the sensitivity shown in your images is set to only 1mV/g, as opposed to 10 - which you listed as the spec for your accelerometers. If this is incorrect, does fixing it change your readings?