Discussion:
Noise issues with PXI-6225 and PXI-6070E
(too old to reply)
Roy-IHRC
2008-07-24 00:10:07 UTC
Permalink
We are using two different DAQ cards (PXI-6225 and PXI-6070E) mounted
on a PXI Chassis to acquire data from different types of sensors.  We
had a LabView application built for this purpose.  We are able to
change the channel configuration depending on the type of sensors we
want to use.  On the current setup we intend to use 64 differential
pressure tranducers from Setra, model 265, which have a 0-5 VDC output
and use a 24 VDC excitation.  These transducers use the changes in
capacitance on the circuit to translate it to changes in pressure.
All channels are configured as NRSE and the first configuration
consisted of all 64 single-ended channels connected to a SCB-68
connector block and then to the PXI=6225.  When we tried to connect all
64 sensors we found noise problems on the signals we acquired, so we
started troubleshooting.  We got 48 of 64 sensors to read acceptable
data with reduced levels of noise.  When attempting to connect the
remaining 16 channels none of the configurations seemed to work.  All
the shieldings in the cables were referenced to the common and still
noise didn't drop on those channels. 
Then we decided to connect those 16 channels through the other card,
PXI-6070E.  Noise levels were reduced, but we noticed that some of the
transducers showed offsets in the "zero".  We recalibrated the sensors,
double-checked the settings, and once we connected them back to the
system, the 0.2V-0.4V offset appeared again.  Also we noticed that in
the previous 48 channels two of the sensors showed considerable levels
of noise.  We tried swithching sensors, checking the cables, shielding,
location on the test specimen, and still can't figure out the source of
the noise.
Talking with the company that developed the LabView application, they
mentioned that the DAQ card may be generating the noise issues.  How
can we confirm this and is this a possibility?
Any feedback is welcomed.
Thanks,
Roy.
tnek
2008-07-24 21:10:24 UTC
Permalink
Hi Roy, You mentioned that you are using the NRSE configuration. I was wondering if you are using the AI Sense terminals (pin 62 on Connector 0 and pin 56 on Connector 1) as recommended in the <a href="http://digital.ni.com/manuals.nsf/websearch/B53FA1760200AF86862573A3006F9E56" target="_blank">M Series User Manual</a>. I would also recommend making sure that all of your connections are secure. <a href="http://zone.ni.com/devzone/cda/tut/p/id/3344" target="_blank">This</a> article has some great advice on reducing noise in measurements. When you mention that you tried swapping sensors and channels, and cables, did the noise seem to follow a particular piece of hardware? Let me know how things go. Regards,KentApplications Engineer
Roy-IHRC
2008-08-01 00:40:21 UTC
Permalink
Hi Kent, yes we've used the AI SENSE terminal on the right pins.&nbsp; I already had a copy of the article you are mentioning, and actually it has helped me to learn a lot about noise and wiring and how to troubleshoot our problem... but still without succesful results.&nbsp; We've been able to get rid of the noise from the PXI-6225.&nbsp; The sensors connected to the PXI-6070E still had some issues and we were able to replace those channels with inputs on an additional brand-new PXI-6225.&nbsp; We are now having problems with the new card.&nbsp; After analyzing the power spectra we determined we are getting noise with a frequency around 9 Hz.&nbsp; It is not coming from the sensors or cables, since we shorted the signal and AISENSE terminals and still got the 9Hz noise on the time history we acquired.&nbsp; This low-frequency noise affects our measurements because we are measuring wind-induced pressure fluctuations which are generally below 10-15 Hz.&nbsp; Anything above that frequency range, we can filter it out with software filters.&nbsp; We are running out of ideas and we don't what else to try or if there is no other option than to double the channel inputs available and go for differential channels.. which at this point may be out of our possibilities.Thanks, Roy Liu.Graduate Research AssistantInternational Hurricane Research CenterFlorida International University
Bueller
2008-08-03 15:40:19 UTC
Permalink
Hi Roy,
&nbsp;
If I may ask, what is the sampling frequency you are using? and would you still get that 9 Hz noise when only one channel is connected?
&nbsp;
Regards,
Faris

Loading...