LabVIEW

cancel
Showing results for 
Search instead for 
Did you mean: 

Uncertainty calculations for cDAQ module NI 9203 and 4-20 mA transducer.

Solved!
Go to solution

Could anyone help me confirm just these calculations for accuracy/uncertainty? I just want to know If I making sense.

 

I have a 16 bit module (NI 9203) that is configured to read 4 to 20 mA signals. That module is connected to a 4-20 mA pressure sensor that reads 30k psi, whose total full scale accuracy is 0.25%. (It goes from 0 to 30kpsi)

 

  • So the accuracy would be ± 75 psi. Right?

 

I was using this formula to calculate the uncertainty of this setup (Ignoring noise for now):

 

  • Uncertainty = (Input Signal x Gain Error) + (Range x Offset Error)

 

If we assume an uncalibrated, and unipolar accuracy (0 to 20 mA) at ambient temperature, the datasheet gives you these values:

 

  • Gain Error: 0.49%
  • Offset Error: 0.46%

 

The datasheet says the real range is 21.5 mA, so would I be correct to assume that for these values, when reading 10 mA, then:

 

  • Accuracy = (0.0049 x 10 mA) + (0.0046 x 20 mA) = 10 ± 0.1479 mA 

 

The pressure sensor has a range of 4 to 20 mA for 30,000 psi. If we assume the sensor really does stop at 0 and 30k, then we'd have 16 mA for 30k psi which is roughly 0.53 uA / psi.

 

  • So for 10 ± 0.1479 mA, we would have 11250 ± 277.31 psi if we do the conversion.
  • Adding up the uncertainty of the pressure sensor: 11250 ± 352.31 psi

 

Am I in the right track? What am I getting wrong, what else should I consider(?)

 

NI 9203 Specifications: https://www.ni.com/docs/en-US/bundle/ni-9203-specs/page/specifications.html

 

0 Kudos
Message 1 of 4
(191 Views)
Solution

Your calculation seems to be correct for a steady-state measurement.

 

Note: All this is "uncalibrated" and you don't operate any critical equipment "uncalibrated"

 

Santhosh
Soliton Technologies

New to the forum? Please read community guidelines and how to ask smart questions

Only two ways to appreciate someone who spent their free time to reply/answer your question - give them Kudos or mark their reply as the answer/solution.
Message 2 of 4
(176 Views)

Thanks, that's what I figured. What would I need to factor in if it wasn't steady state(?)

 

I'm actually just trying to convince a few people to calibrate the modules along with the sensors. They usually just calibrate the sensor itself.

0 Kudos
Message 3 of 4
(151 Views)

The input bandwidth is 850kHz, which means an 850kHz 10mA input signal will read as 7.07mA beyond 850kHz the signal is no longer acquired true to the actual value.

 

This is not a concern if your application does not work with a high-frequency signal.

 

Yeah, definitely you need to educate them on the importance of calibrating everything in the signal chain.

Santhosh
Soliton Technologies

New to the forum? Please read community guidelines and how to ask smart questions

Only two ways to appreciate someone who spent their free time to reply/answer your question - give them Kudos or mark their reply as the answer/solution.
Message 4 of 4
(129 Views)