Signal Conditioning

cancel
Showing results for 
Search instead for 
Did you mean: 

Displayed analog input drifts

I'm using some SCX-1520 channels to monitor voltage inputs.  (Other channels are used for bridge inputs, and they work well.)  When the vi first starts, it displays the correct voltage.  After about 5 seconds it drifts about 0.2 V and stays there.  For example, with 10 V appied, it indicates 10.000 V when the vi first starts.  After about 5 seconds, it displays about 9.8 V and stays there with 10 V still applied.  Sometimes when I electrically touch the analog input, the display will pop back to 10.000 V and then it starts to drift again.   (I say "electrically touch" because I know it's not caused by a bad connection. )  When using MAX to test the system, the displayed value is stable, which suggests to me there something about the programming.  Note also that I'm having this problem with only the channels that I'm using for voltage inputs.
 
The channel property nodes are setup as follows:
 
AI.Lowpass.CutoffFreq = 10
AI.Lowpass.Enable = True  (Setting it to false does not affect the drift.)
AI.AutoZeroMode = Once
 
Any ideas???
0 Kudos
Message 1 of 3
(3,428 Views)

Not sure, but it looks like a bias current problem in your connection.

A (more or less very) small current flows out of your input channel. One solution would be to add a (100k to 10M) R from your input to ground. Search the manuals for bias current. (Sorry, I currently don't know a link, but maybe someone else does 🙂  

Greetings from Germany
Henrik

LV since v3.1

“ground” is a convenient fantasy

'˙˙˙˙uıɐƃɐ lɐıp puɐ °06 ǝuoɥd ɹnoʎ uɹnʇ ǝsɐǝld 'ʎɹɐuıƃɐɯı sı pǝlɐıp ǝʌɐɥ noʎ ɹǝqɯnu ǝɥʇ'


Message 2 of 3
(3,416 Views)

Thank you for your reply.  This is the problem and I should have recognized it.  To make matters worse, the voltage source is an isolated battery powered device.  The link to a paper that describes this issue (common mode rejection) is

http://zone.ni.com/devzone/conceptd.nsf/webmain/01F147E156A1BE15862568650057DF15

specifically, the section titled "Measuring Floating (Nonreferenced) Sources"
 
Again thank you,
 
DJ
0 Kudos
Message 3 of 3
(3,409 Views)