NI 9205 noise and random noise spikes

[sfeltz]

I'm using a single NI 9205 module in a cDAQ 9178 chassis with a number of other modules including 3x NI 9229, 1x NI 9213, and 2x NI 9472.  The 9205 is being used to sense the current across a shunt resistor (150A, 100mV resistor) in differential mode.  At very low currents, such as ~0.15A on a 9V battery, which would be about 100uV across the resistor, we are seeing some noise, such that it fluctuates around 0.18A to 0.12A or so.  Given the sensitivity of 4uV and noise level of 10uV at the +/- 200mV scaling, we would expect to see about this much difference, correct?  As far as I can tell, there is no way to improve the sensitivity.

Secondly, when the short on our circuit is removed so that we have an open circuit, the current still gives some minor readings of around -0.054A to 0.052A, but it also randomly spikes to -20A to -30A.  In terms of connections, I have the positive to positive, the negative to negative in the differential pairs on the card.  The COM port is connected to a 24V supply Gnd (has also been tried at earth gnd to no change).  The one thing that confuses me in the manual is it shows the negative connections tied together and attached to COM externally.  I can't tell if this is supposed to happen outside the card or if it occurs internally, because it seems like if its happening externally, then it basically becomes an RSE set up and not differential.

If there is any solution to this, please let me know.  Thanks.

[margasan]

Hello sfeltz

Unfortunatelly there is no way to improve the device sensitivity, as you are already using the smallest range. The random spikes could be the result of some enviromental electromagnetic noise, have you tried moving the device to a different location?? Can you post a diagram of your actual connections and the page of the manual you are looking at??

Regards

Mart G.

[sfeltz]

Thanks for the reply margasan.  I've actually been working with an Applications Engineer at NI through email to figure this out.  It seems we had a power supply a little too close to the chassis, the shielded signal wire wasn't grounded and we also needed bias resistors, which the manual doesn't mention, only the field wiring guide. Things work a lot better now.