NI 6255 accuracy calculations

[JLAnderson]

I am trying to determine the frequency and voltage accuracy of a RS-485 output signal running at 115K baud.  The 6255 specifications state the timing accuracy is 50 ppm of sample rate.  Taking 50 ppm of the period will give me a small error but the zero crossings may fall between sample points.  I could calculate the slope of the zero crossing but there are few samples to do it accurately.

For voltage the assumed basic accuracy is 10v/16 bits but the settling time may be a significant factor.  Not sure of what a full scale step is.  Two channels are measured at 750KS/s alternating between 3 and -3V volts.

Any suggestions?

[Jason-M]

JLAnderson,

The closest full scale range with that signal would be between -5 and 5 V. You could also use +/- 10 V, but it will have half the resolution. Have you checked out the Datasheet? This has a great overview of all of the aspects that are involved in absolute accuracy. For your reference, you can ensure you understand our jargon HERE. Also, if you do not want to do the absolute accuracy calculations yourself from the datasheet, you can plug the values into THIS example. Let me know if this provides enough information to find what you are looking for.

Best,