The NI 4071 DMM accuracy specification on the one volt DC range for 2 year (18-28) is 15+0.8. Using a shorting plug in the input for 0 volts means the accuracy is ((0*15)+(1*0.8)*1E-6) = 0.8 uV. The calibration procedure specifies 2.1 uV (Table 13). The calibration specification must be right since the 100 mV accuracy also calculates to be 0.8 uV. Please explain the discrepancy.
Solved! Go to Solution.
The uncertainty on the 4071 measuring 0V in the 1V range is ±0.8uV for two years in an ambient temperature between 18 and 28°C within 1°C of Tcal.
Looking at table 21 (DC Voltage Verification Limits) of the 4071 calibration procedure, we see that the 24 hour uncertainty verification limit is ±0.4uV, and that the 2 year limits are ±0.8uV.
I was looking at cal spec dated Apr 07 but an now looking at one dated April 11. Table 21 is for AC volts while Table 20 is for DC volts. It look like you are on the 100 mV range instead of the 1 V range that shows 2.1 uV. All the entries in the DC Accuracy Specifications match the range times the second number except on the 1 V range. We are calibrating on a 2 year cycle so we are using the 2 year numbers.
Apologies, yes, I referenced the 100mV range instead of the 1V range, and yes, I meant to write 'table 20', not 'table 21'. Let's start over :).
The specified uncertainty on the 4071 measuring 0V in the 1V range is ±0.8uV for two years in an ambient temperature between 18 and 28°C within 1°C of Tcal.
Looking at table 20 (DC Voltage Verification Limits) of the 4071 calibration procedure, we see that the 24 hour uncertainty verification limit is ±2.1uV, and that the 2 year limits are ±2.1uV.
Now we're on the same page.
I agree that it appears the uncertainty limits for calibration are wider than the guaranteed uncertainty listed in the specifications. Let me look into this more and determine what we're missing.
The specified uncertainty for measuring 0 V in the 1 V range is wider in the Calibration Procedure because offset nulling is not performed during calibration. Per the 4071 specification, you must add 1.3 ppm to the range when not performing offset nulling. Therefore your 24 hour uncertainty is [4 ppm * 0] + [(0.8 + 1.3) ppm *1] = +/- 2.1 uV. The specification derating is captured in the 5th footnote in the DC Voltage Accuracy table in the PXI-4071 Specifications document. I've pasted it below for reference.