usb-6218 intermittent signal

[Jim12345678]

I'm using a usb-6218 to read a differential analog input.  I've done this hundreds of times now for years.  I'm using a 1.5V battery to test the input and Ni-Max just to trouble shoot.  The signal is intermittent.  It jumps to 1.5V on occasion and then back to 0.  The connections are all good.  This happens regardless of what analog channel I try.  Any ideas?  Any idea how to tell if the DAQ is just broken?

[will.i.am10]

Hello Jim,

Have you tried testing it with anything else besides the battery? Such as using a power supply, or the analog outputs on the DAQ device? The link below outlines how to do an external loopback test, this should give you more information into if the card is actually reading incorrectly.

Loopback Test: http://digital.ni.com/public.nsf/allkb/6656E931364BB6F18625765400409DE1

Thanks,

Will

[Jim12345678]

I'll give it a shot, but I think it has more to do with shielding and cross talk than the DAQ not reading correctly.  I'm using twisted foil shielded wire with a shield wire.  In addition to the analog input, I've wired a counter input.  I can generate the intermittent signal by connecting the counter shield wire to my electronics even if there isn't a pulse train being generated.  Counter channel appears to work fine.

[will.i.am10]

Hi Jim,

Have you tried with other wires? I am curious as to why you are seeing so many lines during each intermittent signal read. I would expect only one line, not all the other lines and colors within each pulse. Did you have any interesting results when doing the loop back test?

If you think this is because crosstalk I have attached a document with troubleshooting steps to reduce this.

Troubleshooting Crosstalk: http://digital.ni.com/public.nsf/allkb/B9BCDFD960C06B9186256A37007490CD

Have a good day,

Will

[Jim12345678]

I'm reading multiple channels but am only applying a signal to one line.  I've found that adding a 1kOhm resistor between the differential return line and the DAQ analog ground stabilizes the signal for all inputs (even though I just add the resistor for one channel).  Next I have to figure out why wiring up a digital input line causes the analog lines to go haywire.......