Our company makes a commercial instrument and Ive recently found that we have a ground loop passing through a NI card. The device has been around for almost 10 years in its current form although this problem is probably new in nature or discovery. We are using a very old NI card (NI 6025E) in our current production but we will be updating sometime soon. There is also a NI6503 within the system but does not appear to be related to the ground loop.
Let me explain the basic of the hardware. We have probe which can move two motors to direct the position of our instrument. The probe is on the end of 10ft long probe cable. These motors are primarily controlled by a Galil controller and Galil Card.
The NI card directs some data to the to the probe (not related to the motors) , including sensing of the ground plane of this cable (presence of 0V means probe connected) (ideally Galil senses this same signal too). The majority of what the NI card does is in the custom electronics unrelated to galil controller and motors.
So the galil card generates a pulse to indicate motor position (and Im getting support with Galil too).. The pulse should be from 5V and changes to 0V to indicate a trigger event occurred. Where the problem is that the pulse we are getting has a different voltage, ie, 4.5V to -0.5V when measured across the CMP signal to ground. Ive measured an offset on some devices from 0.4V in one device but only 0.06 to 0.05 in others. This CMP pulse comes directly from the galil card into a PC board that interfaces with the cable.
The funny part is that when I disconnect the NI cable from the 6025E, the signal goes to 0 to 5V. Im not sure I can really consider this a NI problem but any ideas or help is appreciated.