Lab Equipment Grounding

Thread Starter

blah2222

Joined May 3, 2010
582
Hello,

I work in a lab that measures small physiological electrical potentials and the topic of equipment grounding has been of interest to me. At times, mains noise appears at large amplitudes and intermittently which completely swashes out the signals of interest. Also, small DC offsets appear even when our amplifiers and DAQ have high pass filters enabled.

I have attached a physical schematic of how my equipment is internally connected to ground. All pieces of equipment are powered via powerbar which is connected to a single wall socket.

The outputs of each amplifier is connected (via BNC) to a separate channel of the DAQ. Each amplifier's AMP GND is connected (via banana plug) to one another and then connected to a large metal table where the measurements are made.

I am curious if this looks to be noise prone either due to common references or ground loops. Would it be recommended to connect each amplifier's CHASSIS GND to AMP GND (as shown with dotted lines)?

**The resistors (49.9 Ohm & 100 Ohm) are internal to the devices.

Thank you for your time
 

Attachments

Last edited:

MrChips

Joined Oct 2, 2009
30,706
Ground loops are one of the most difficult things to eliminate when working with very low level signals.

The first thing I would do is separate SIGNAL GND from CHASSIS GND.
For example, don't connect your metal table to any lab instrument. Ground the table to the AC service GND.

For very low level biological signals, you need amplifiers that have very high common mode rejection ratios (CMRR) which you will get if the amplifier uses proper instrumentation amplifier designs (INA). The inputs are usually three leads, a +ve lead, -ve lead, and GND.
 

MisterBill2

Joined Jan 23, 2018
18,167
I suggest removing the connection between the amplifier ground and the amplifier output going to the Data acquisition system. If the amplifiers have truly differential inputs, then the resistors are probably not doing you any good either. Likewise on the DAQ system. Connecting a ground creates a point that can deliver an input is my reasoning. The connections from the table to the amplifier grounds are needed to keep the input common mode voltages in line, so keep those connections.
 
Top