USB-6008 analog input is unstable

[zener]

I am connecting a pressure transducer to the USB-6008 .   I am using differential mode and trying to read from the NI-MAX test panel.  The readings vary with the pressure but are extremely unstable.

The reading is very unstable even when I select "single sample".   Although the actual voltage across the resistor is quite stable.  The problem seens to be internal to the USB-6008.  Any suggestions?

[zener]

Update,  After trying several different methods of connecting the signal, I finally found out what was causing the interference.  I had to place a 10 uF 25V capacitor across the inputs of the Channel.  This eliminated the unstable signal. 

Must be coming from internal in the USB-6008.  I even tried using a Fluke 787 to source a signal and the USB-6008 was still very unstable.  The capacitor fixed it.

[devchander]

Try single ended mode once

Or, Is your measurement unstable in Single ended mode too?

[zener]

Hi, Thanks for the respose.

I did try single ended.  Still the same instability.  Seemed slightly better in differential although not much.  The capacitor worked perfectly.  The fact that it was still noisy using the fluke, and with all other power supplies off,  indicates to me that the noise is originating from the USB-6008 itself.  Does this seem feasable?

[herrit]

hi,

I had problems too by measuring a voltage with the analog input of the NI USB-6008. I use differential as configuration with the virtual channel vi. after 8 minutes of stable singal it becomes verry unstable en noisy. There is a screenshot in the attachment too show you the problem.

Thanks in advance

Gerrit

[herrit]

hi,

does anybody knows how to solve this problem? I've been running it again a few times and sometimes it happens only after half an hour. The noisy signal happens also often in the beginning of the measurement. But I can't find an explination why ... what am I doing wrong?

[AnalogKid2DigitalMan]

What is the thermocouple connected to?

Are there any electrical devices or machines that turn on/off in the vicinity that could be generating electrical or radiated noise?

[herrit]

dear,

The thermocouple is directly connected with te daq. There are no machines in the vicinity that might disturb the signal. I was thinking about a bad PCB contact but i couldn't found one yet. Like said before i'll try a capacitor but if that was the issue there would be a bad signal all the time.

[zener]

Hi,

I posted the original problem with this.  Here is what I found.  I added the capacitor and it fixed the problem, however, there was another "smoking gun" that was the original problem..  Open the Measurement and automation explorer (MAX) and select tools>nidaq mx configuration>ni-daq mx usb driver selector.  See if you have two choices there such as Ni-daq MX and Ni-daq MX base.  If you do, try the other one.  This was causing the unstable readings I encountered.  After changing it, I no longer needed the capacitor.

I had both drivers. Apparently one came from the real-time or PDA software.

Hope this helps. 

[PBear]

Hi all,

I'd like to point everybody to this knowledgebase. I know many times customers have some problems with readings becoming noisy after awhile and has to do with charge build up on a floating signal. Sometimes customers claim that after simply touching their device (for example, thermocouples) they get good readings again. This is because the built up charge now has a place to discharge. So read this and it should explain why a capacitor helped (which may have just filtered the signal more than providing an AC path to ground.)

KnowledgeBase 11TARTWV: Why Do I Need to Ground My Thermocouples When I Am Already Measuring Them in Differential Mode?
http://digital.ni.com/public.nsf/allkb/862567530005F09C86256522005CD1EF?OpenDocument