I'm measuring the signal from a mass air flow device (Hot wire-type measurement) that outputs a 4-20mA signal using an NI9203 module in a cDAQ 9178 chassis. I'm measureing the signal at a 10Hz sample speed. The signal ground, chassis ground, and common terminal on the 9203 all go to the same location. The sensor it self is not grounded, but is powered by the same 24VDC power supply as the current loop. I'm getting some abnormal spikes (noise) in my measurement and was hoping someoen could provide some direction.
I've attached a little .pdf snippet that shows the signal. The variation in the bunch in the middle is expected (upwards of 50 scfm variation is within our expected std. dev.), it's the large spikes, especially the one centered in the 150 second area and the like that have me concerned. I haven't gotten out the oscilloscope yet to determine frequency of the spikes, but am planning to do that tomorrow. Just wondering if anyone had any thoughts or ideas prior to that?
Thanks for any help you can provide,
This is Andrew Brown, an Applications Engineer from National Instruments. It's difficult to say the source of your noise offhand. Some good resources that should help you work through reducing noise are the following articles:
Please let me know if you have questions about these articles or about the process of troubleshooting the noise in your system. Thanks, and have a great day!
All very good articles. In the end software fixed it though. By applying a low-pass butterworth filter we cleaned the signal up significantly and stayed well below the response time of our sensor, so a win win.
Thank for your help.