Accuracy Issue (NI 9216 & NI 9217)

[ckc8302]

Gurus,

I am having a custom test apparatus built by an outside contractor that is to include an NI cRIO with a few different DAQ modules; including the NI 9216 and NI 9217 RTD modules. Our uncertainty (accuracy) requirements for air and liquid temperature measurements (RTDs) are 0.15°C and 0.2°C, respectively. After looking at the data sheets for these modules, I am concerned in that they show temperature accuracies of 0.15°C for the module alone (4-wire mode). This appears to leave no room left for uncertainty in the RTD itself. I assume that we need to account for error-stacking between the sensor itself and the DAQ module. Am I incorrect in my reasoning, is there a more accurate DAQ card that can be specified, or are we simply not going to be able to meet the specified accuracy requirement? Thank you in advance for any insight that you all may have.

[yamadj · Accepted solution]

Hello ckc8302,

Thanks for posting your question as it always a good idea to ask and make sure devices meet specification.

Looking through the datasheets for these two devices it looks like the modules themselves are not going to be able to meet your spec.

NI 9216 Datasheet - http://www.ni.com/pdf/manuals/376921c_02.pdf

NI 9217 Datasheet - http://www.ni.com/pdf/manuals/374187a_02.pdf

In addition, I've already tried looking throug our other c series modules and other hardware listed in the article below.

Choosing an RTD or Thermistor Measurement System - http://www.ni.com/white-paper/3643/en/

The article also lists the 9226 and the 9219 but both still don't seem the meet your spec. 

It looks like the only card capable would be the NI PXIe-4357 but at this point you may have to reconsider your system configuration.

NI PXIe-4357 Data Sheet - http://www.ni.com/datasheet/pdf/en/ds-413

Hope this information was helpful!

[ckc8302]

yamadj,

Thank you for your detailed reply with references. You definitely answered my questions as I asked. I think that we may actually have been dealing largely with some confusion with regard to accuracy vs. uncertainty. There is a European standard EN14511 that requires certain "limits of uncertainties"; which is really a different (and much more complex) metric to quantify.

I will have to do some more research into this to come up with a concrete answer, but I am considering this issue answered and I thank you again for your help.

[yamadj]

Hey ckc8302,

I'm glad I could help! Thanks you for posting the question, I'm sure others will benefit from it as well!