2
\$\begingroup\$

I have an accelerometer (AC208-1D by CTC). This accelerometer could give proper readings when I would shake it in my hand (the right frequency and the right gs). However, once I turned on a motor (3-phase motor that is connected to a 1-phase to 3-phase converter plugged in to the wall), the accelerometer could no longer read inputs from me shaking it, even though the excitation that was read when I shook the accelerometer was greater than the excitation from the noise.

This is the bare signal (no filters) from the accelerometer while it's placed where it needs to be for an experiment, and the motor is off: enter image description here

This is the bare signal with the accelerometer in the same place and with the motor plugged in to the same outlet as the accelerometer's daq, and the motor has power supplied to it (motor not running): enter image description here

And this is the bare signal with the motor plugged in to a wall outlet about 40-50ft away from the daq's outlet with the motor still receiving power (motor not running): enter image description here

The noise caused by current being supplied to the motor seemed to be the same, just with the thinner spikes going in different directions. I thought that the noise could be generated just from the motor's EM field since it is a .5HP motor, but based on my measurements of the EMF, I think that this isn't really what's causing the problem.

This is the EMF measured in microTorrs of my phone sitting on top of the motor while it is off: enter image description here This shows a large peak when I turn the motor on and it stays constant until I turn the motor off where it receives another peak: enter image description here

However, the following two measurements measure where the accelerometer sits. This first picture is my phone's reading while the motor is off: enter image description here Then, this picture shows when I turn the motor on at about 9-10s and off at about 15-16s. The large dip shown past 20s was from me picking up my phone. enter image description here

So, from the information above, I don't think that the problem has to do with the motor producing an EMF when it is receiving current. The wire connected to the accelerometer has foil and braided shielding, and the shielding is attached to the daq's ground, and this wire is far away enough from the motor that the motor's EMF should be negligible. I was wondering what I could do to remove this noise caused by the motor just existing with current flowing to it.

After working at it, I figured that the problem was that my DAQ is reading the current that flows back into the DAQ through the neutral wire. My DAQ is powered with DC power. If I were to throw in a Schottky diode or a reverse polarity diode, would I be able to eliminate the problem where the DAQ overwrites my accelerometer's data with the voltage data it gets from the reverse flow? I know that these diodes draw voltages around 300mV to 600mV, but the readings from my accelerometer will be in the 400mV range, and I don't know if the voltage draw from the diodes would affect the data read from the accelerometer in a significant manner.

Also, it turns out that there is a disturbance caused by the motor's EMF. The air has currents of about .1A when measured by a multimeter, and the current between the accelerometer fluctuates between .7A to 1.7A.

\$\endgroup\$
7
  • \$\begingroup\$ I bet the DAQ doesn't use balanced differential inputs? \$\endgroup\$
    – Andy aka
    May 12, 2020 at 16:41
  • \$\begingroup\$ The DAQ is an NI9230, and I think it does use that. How would I go about confirming that it does? On the datasheet, it says it has an input impedance differential of 324kohm \$\endgroup\$
    – Albert Garcia
    May 12, 2020 at 16:46
  • \$\begingroup\$ Are you using a proper cRIO interface then? \$\endgroup\$
    – Andy aka
    May 12, 2020 at 16:52
  • \$\begingroup\$ I think so. I have the DAQ connected to a cDAQ-9188 chassis, and that goes to a labview program that translates the voltage to gs using the sensitivity of the accelerometer. \$\endgroup\$
    – Albert Garcia
    May 12, 2020 at 16:58
  • \$\begingroup\$ I plugged in the motor to a different building using a bunch of electrical cords, and it worked out. I am guessing the ground of the electrical motor interferes with the ground of the accelerometer, and this created an interference in the signals. Is there an easy way to fix this ground loop interference then? There is interference when we use other high-voltage equipment as well such as a water cooler and welder. \$\endgroup\$
    – Albert Garcia
    May 12, 2020 at 18:15

1 Answer 1

Reset to default
0
\$\begingroup\$

I separated the motor from the steel table it was mounted to with some wood, and I also grounded the motor to a different ground from the DAQ's ground. Some interference still existed after this step, but the accelerometer's data was no longer being overwritten by the motor. The accelerometer will be mounted differently to the test fixture. The threads of the accelerometer being screwed in to the test fixture demonstrated a clear increase in noise. The more contact the threads had with the fixture, the more noise that was presented.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.