Multifunction DAQ

cancel
Showing results for 
Search instead for 
Did you mean: 

noisy measurment

I am using a PXIe-6363 to measure small amount of voltage through Analog inputs. suprisingly I am seeing alot of noise, even though I used shielded wires and NI terminal box. I was using the cheap USB 6008 before and did not have this problem, and decided to improve the system resolution and speed by using the PXI card. 

 

anybody have any suggestion how to elimiate the noise the system is picking?

0 Kudos
Message 1 of 4
(3,908 Views)

It really depends on what is generating the noise, how large the noise actually is, what the requirements are to adequately measure the desired signal components, how the noise is getting into the system, and many other factors.

 

Please tell us what you are measuring. How is the desired signal generated? What are the sources of the noise or interferring signals? How large are the desired signals and the noise or interference? What are the frequency ranges of the signals and the noise? What will you do with the data after you measure it? How fast do you need measurement updates? How is everything grounded?

 

Lynn

0 Kudos
Message 2 of 4
(3,886 Views)

The system is using a simple NI USB-6008 DAQ to convert the Analog voltage measurement to digital. It is a 12 bit device with 37.5 mv Accuracy and we attached it to the back of plating board with normal wires. We see acceptable measurements with almost no noise. During the improvement we liked to have a higher resolution of our reading plus more analog input we replaced the 6008 with a more sophisticated PXIE-6363 Multifunction DAQ which has 16 bit resolution and 33uv Accuracy. The card sits in a PXI chassis 1 foot from the plating board. Unfortunately the measurement has a lot of noise using the PXI.

 

We tried to identify some sources of noise. I replaced the cables that connects the plating board to the PXI with coax 50ohm 26 AWG cable (A307-100-ND) and replaced all the ground cables that were hooked to the pins, with a plug connection type. I observed a very small improvement but still a lot of noise. The other possible source of noise might be the movement of cables from plate to the ADC, which is attached to the Robot arm. although we have the same movement with USB-6008 , since it is attached to the plating board which is fixed to the Robot arm, but the wires are much shorter and probably won’t move as much.

 

Our measurement is in 0.1~0.6V range with 5kHz sampling rate:

 

 

 

 

 

 

 

 

Download All
0 Kudos
Message 3 of 4
(3,869 Views)

Try to find out more about your noise:

Sample one channel only as fast as possible . take 100k to 200k samples and make a power spectrum..... noise is a floor .... do you see spikes?

 

if you sample with 50kSPS and decimate by 10 with mean ... what do you get?

Greetings from Germany
Henrik

LV since v3.1

“ground” is a convenient fantasy

'˙˙˙˙uıɐƃɐ lɐıp puɐ °06 ǝuoɥd ɹnoʎ uɹnʇ ǝsɐǝld 'ʎɹɐuıƃɐɯı sı pǝlɐıp ǝʌɐɥ noʎ ɹǝqɯnu ǝɥʇ'


0 Kudos
Message 4 of 4
(3,854 Views)