Making many psuedodifferential measurements without loading ground

[peabody124]

I'm recording many differential signals with a fairly long wire (25 m).  The negative of all of them is connected to the PXI ground through a cable.  I bring back groups of 28 signals with a ground reference and the 28 signals in a single cable.  Currently I have the ground reference go into my 28 negative inputs on the 4498, and the other side go into the positive.  However, because there are 28 50 ohm connections to ground, the end result is each of my groups of 28 create a ground loop.

I'm wondering for recording something like this if the right way to do it would be to send out the PXI ground to the negative of all of them, then feed the 28 + inputs into 28 positive channels, and the ground reference into another positive channel.  Then I could let the - inputs to the 4498 pull themselves to the PXI ground.  Does this sound sensible?

Thanks!

[ManLongAtNI]

What device are you measuring? It sounds like you have a grounded signal source. If this is the case you should be good connecting the negative/ground terminal of the source to the negative terminal of the 4498. Please use the

Field Wiring and Noise Considerations for Analog Signals DeveloperZone article as a reference to connect your system accordingly.

Regards,

Glenn

Applications Engineer

National Instruments

[peabody124]

I'm measuring a physiological signal, so there is a unity gain preamplifier at the source, so I have essentially a copy of the ground reference the preamplifiers use.  The issue with using the negative input, is because of the pseudodifferential nature of the PXI-4498, having a bunch of negatives wired to the chassis ends up creating a very low impedance on my ground reference (from the source) to the chassis.  Because there are a handful of copies of that ground reference (for different preamplifiers) they have a slightly different parasitic divider on the ground and I get a ground loop.  I've read that developer note, but unfortunately it doesn't address the fact that a ton of pesudodifferential negatives ends up being a short to ground instead of a high impedance connection.

[ManLongAtNI]

Can you provide a wiring diagram of you current setup with the 4498?

Regards, 

Glenn