Quantization Error for NI AT-MIO-16E-2

[karthik_sampath]

Hi
I wanted to confirm the value of quantization error for the 12-bit NI AT-MIO-16E-2 DAQ. I can take values between -10 to 10 Volts. So the quanitization error would be:
= +/- 0.5*(20/(2^12)) = +/-2.44e-3

Is this correct? This seems like a high amount of error when voltages are below 0.1V

Also, I want to know if the above DAQ card comes with an anti-aliasing filter, depending on the sampling frequency.
Thank you
Karthik

[LYu]

Hi Karthik,

If you choose your measurement range to be +/-10V, the smallest change that can be detected is 20/2^12 = 4.8mV. So the maximum quantization error would be +/-4.8mV. If you're looking to increase the accuracy, you can change the measurement range to be the smallest possible for your requirements.

There's no anti-aliasing filter built in to the DAQ card, but you can easily implement them on a terminal block (analog) such as the SCB-68 or in software (e.g. Low Pass Filter in LabVIEW).

Analog Input Low Pass Filtering with an SCB-68 Terminal Block:
http://digital.ni.com/public.nsf/websearch/5D82A2990FB1D21386256EEE00018FF0?OpenDocument

Hope this helps!

Thanks,
Lesley Y.
Applications Engineering

[karthik_sampath]

How do I change the measurement range? Can I do this using LabView or should something be changed at the DAQ itself?
Thanks
Karthik

[LYu]

Hi Karthik,

You can configure the channel input limit in LabVIEW using the AI Config VI. The available measurement ranges can be found at the following link:

[LYu]

Sorry, posted too quickly. Here's the link to the specifications for the AT-MIO-16E-2:

href="http://www.ni.com/pdf/products/us/2mhw239-240e.pdf"

[Dennis_Knutson]

You can certainly do filtering in LabVIW but that's not the same as an anti-aliasing filter. No amount of software filtering will rid your signal of aliasing effects.