Multifuction DAQ keeps disconnecting while welding or motor accelerates rapidly

[muharani2014]

I am using multifunction DAQ X series USB. In the workshop, when welding is conducted the DAQ keeps disconnecting. Also, as I am using this DAQ to give RPM setpoint to a motor drive, when the motor ramps up with the acceleration of 5 kRPM/s, the same disconnection happens. When the acceleration is set at lower values, say 1 kRPM/s, this problem does not occur. I think this might be due to an impluse going through the ground since both DAQ and the drive are grounded to the test rig and IO's on the DQA are connected in RSE configuration. Is there anyway to isolate the DAQ ground for example putting a capacitor (series with a small resistance or without it albeit not healthy!) between the DAQ ground wire and the test rig?  If the solution involves connecting them in differential configuration I would run out of the terminals.

[johnsold]

It certainly sounds like a ground loop could be the problem.

The ideal solution is for each device (computer, DAQ, sensors, welder, motor controller, motor, ...) to be connected to the same ground point exactly one time. Unfortunately, this is often impossible due to internal connections in various devices.

The first thing to do is to draw a very accurate diagram showing all known and suspected connections to ground. The USB ground is connected to the computer ground through the cable. Note also that there may be more than one "ground" in the system. Circuit common or return may be connected to chassis or earth ground or not. The power system ground is intended to protect the power system (first) and equipment and people (second) from the consequences of faults. As a result this ground is often a poor reference for signals.  The National Electric Code allows 10 volts across grounding conductors under fault conditions! Anything with this much impedance is a poor grround from an instrumentation perspective.

After you have a good document showing the connections, then you can try to see where unwanted currents may be flowing and what to do about them. Also consider that you may have inductive coupling into a ground conductor from the high currents of the motor or welder even if the grounds are properly connected.

This can be a very challenging problem. Sometimes simple re-arrangements of grounding conductors is sufficient. Other times significant redesign using isolated sensors, DAQ equipment, or computers may be required.

Lynn