NI 9220 waveform Glitch

[ruthlcw]

Hi,

I have a test setup that measure 16 voltage inputs and 16 current inputs. All current signals go thru a transducer to convert into voltage signals. All 32 voltage inputs are send to a PXI system and a cRIO (redundant RT monitoring system) using differential measurement. I validated the PXI system is measuring the inputs correctly and waveforms look good.

For the cRIO system, I have all 16 voltage inputs go to one NI 9220 and the 16 current signals go into another NI 9220. I am using the FPGA to acquire the data and send to RT via the DMA FIFO. In the RT, I am reading the voltage signals correctly from one of the NI 9220, but the current signals on the other NI 9220 are showing waveforms with glitches.

Prior to energizing the test setup, I used a benchtop power source equipment to inject and voltage and current signals to validate my cRIO program, the measurement and waveforms show a nice and smooth sinusoidal waveform. however, when I energize the setup, the NI9220 that receive the current signals are not displaying correctly. I know my CT transducers are working properly because the signal send to cRIO also send to the PXI and the PXI measurement show pure sinusoidal waveforms.

As part of the troubleshooting, I swapped the NI 9220 module, added a 440kohms from the AI- to COM per the NI documentation, but none of these works. I am wondering if anyone have any idea or recommendation that may cause this glitches appear on the waveform. 

Thanks,

[dsb@NI]

It looks like a ground loop. Make sure you only have one ground in your system. While it was useful to use the PXI system to validate smooth signals, try removing it to see what effect that has on the cRIO measurements. You indicated that the 9220 channels are measuring in differential mode - are you grounding the cRIO chassis?

[ruthlcw]

Doug

Interesting. 

My understanding is that all ground connections go back to one ground bus in my setup however, I just realized there is a ground lug on the chassis. I thought the chassis was grounded through the din rail and backplate is sufficient. I will try to connect a ground to the chassis gnd lug see if this solve the problem. One thing I don't understand is that why does it only affects the 9220 with current signals while the other 9220 is ok. 

[ruthlcw]

Quick update: 

I added GND to the ground lug on the cRIO - No Change

Removed PXI signal - No change