Multifunction DAQ

cancel
Showing results for 
Search instead for 
Did you mean: 

Filter out Device Noise

I am using the NI USB_6001 DAQ. By wiring the DAQ to itself (AI 0+ & AI 4-) and running a differential measurement on the device itself, I am getting a noise reading that switches between  -0.0002 V and -0.0015 V (picture attached). Resulting in -1.3 mV noise being in the device alone. When trying to get a reading of 5V from a power supply I am getting readings that switch between 4.9891 V and 4.9879 V (picture attached). Again that ~1.3 mV switching. 

Is there a way to measure the device noise and filter it out from the voltage read? 

I am trying to read as close to a flat DC line as possible. 

 

Is there a way, perhaps in LabView, to set up something that will measure the internal noise by wiring a short wire from pins AI 3 and AI 7, running a differential measurement on those pins to get the device noise with no voltage being supplied. And then to read pins AI 0 and AI 4 and subtract the device noise from the results of the voltage read? 

 

I have read the Field Wiring and Noise Considerations for Analog Signals a couple of times and already aware that I am using a Grounded signal source (power supply) and a differential measuring system (DAQ). I know that this can cause a Ground-loop noise to be present. I have tried numerous different wires and different wire lengths with no change. I also am aware that there is some miniscule noise variation when my laptop charger is plugged in vs not plugged in.  

 

Any insight would be greatly appreciated.

0 Kudos
Message 1 of 2
(1,566 Views)

Your device supports only a 20V range from -10 to +10 VDC.  It digitizes with 14 bits.  20V / (2^14) ~= 0.0012 V.

 

What you're seeing is 1 bit worth of quantization noise in the LSB of your A/D conversion.  I don't know if you're gonna be able to do better than that.

 

If I *had* to try, I'd oversample and average.   The theory is that as the LSB toggles back and forth, a mean will weight the two values proportionally to the time spent at each.

 

 

-Kevin P

CAUTION! New LabVIEW adopters -- it's too late for me, but you *can* save yourself. The new subscription policy for LabVIEW puts NI's hand in your wallet for the rest of your working life. Are you sure you're *that* dedicated to LabVIEW? (Summary of my reasons in this post, part of a voluminous thread of mostly complaints starting here).
0 Kudos
Message 2 of 2
(1,541 Views)