USB-6211 is wreaking havoc with a voltage divider

[MDSI94]

Hi all,

I'm hoping that somene here might have a new idea, because I'm completely out. I'm trying to use a UBS-6211 board to measure the voltage output from the power source in a system I have set up. The voltage ranges from 0V to 55V, so I've got it hooked up to a voltage divider in parallel to the rest of the system, as shown in the Circuit file. This seems to work fine, as judged by a voltmeter, as long as the board isn't connected. Once the board is connected, all of the readings get strange. The attached Plots file shows the behavior of the system, both according to the voltmeter (DMM) reading the voltage across the voltage divider and LABView (Comp).

The voltmeter makes it look like there's a diode or something of that sort that switches on and changes the impediance of the system, but the read in voltages are such a mess that I really don't know. I've tried a bunch of different combinations of resistors, and it seems that they don't much matter unless they're exteremly large or small, which is expected. One other thing to know is that there is almost no current drawn by the system, about .2 mA, until the elbow, at which point the current drawn steadily increases. 

If anyone has suggestions or would like more information, I'd be glad to hear from you.

Thanks,

-Matt

[johnsold]

Where are the grounds connected? You do not show the USB6211 ground.  The specification for the device requires that all input voltages be within +/-10.4 V of AI GND.  AI GND is usually connected to th eUSB ground which in turn is connected to the computer ground.

If the voltage goes outside the specified limits, some non-linear behavior may occur.

Lynn

[MDSI94]

Oh, interesting. So even if the board is being given a differential voltage to measure, that voltage must be within 10V of ground? Everything is floating at the moment so if that's the case, it may explain the issue.

[johnsold]

Yes.  Absolutely the voltages must be within some limit of ground.  There is nothing magic about differential inputs which would allow them to float thousands of volts (or even 15 V) above ground.

Lynn

[SnowMule]

Definitely grounding and maximum input relative to ground. 

Might consider a differencing amplifier rather than just the voltage divider circuit.