Calibration

Calibration and Temperature Variation

When a system is composed of multiple integrated instruments, the system is subject to temperature rise caused by inherent compromises in air circulation and other factors. Self-heating from surrounding equipment, uncontrolled manufacturing floor environment, and dirty fan filters are among these factors.

Refer to the PXIe-4112 Specifications for the following information for your instrument:

• Recommended operating temperature range
• Calibration interval

Refer to Best Practices for Building and Maintaining PXI Systems for the definition of ambient temperature.

If the ambient temperature is outside of the specified range, you may need to know the measurement accuracy to account for temperature variation. One way to calculate the specified accuracy outside of the temperature range is to externally calibrate the system at the desired temperature. External calibration, though inconvenient, should allow the device to attain its full rated accuracy at the calibration temperature. You can learn more about external calibration at ni.com/calibration.

Another way to calculate the specified accuracy outside of the temperature range is to add the temperature coefficient accuracy for each additional degree outside the calibration range.

The following equation represents the temperature coefficient (tempco).

Tempco = X% of accuracy specification/°C

For example, consider an instrument outputting 5 V with voltage accuracy specified at 0.05% of output + 100 µV in the range 18 °C to 28 °C, and tempco specified as 10% of accuracy specification per °C. If the last external calibration was performed at 23 °C, the following equation represents the 1-year accuracy of the instrument in the 18 °C to 28 °C range:

0.05% of 5 V + 100 µV = 2.6 mV

If the ambient temperature changes to 38 °C, the device is operating 10 degrees outside the specified range, the accuracy is calculated as follows:

±(2.6 mV + ((10% of 2.6 mV)/°C) * 10 °C) = ±5.2 mV

The total error is twice the specified error (5.2 mV in the example above, versus 2.6 mV if temperature effect is ignored) due to the 38 °C ambient temperature. If the additional error term due to temperature drift is unacceptable, some devices support self-calibration at the desired measurement temperature to improve accuracy.

Refer to the PXIe-4112 Calibration Procedure for the external calibration procedure for your instrument.

Accuracy represents the uncertainty of a given measurement or output level and can be defined in terms of the deviation from an ideal transfer function, as follows:

y = mx + b

where m is the ideal gain of the system

x is the input to the system

b is the offset of the system

Applying this example to a power supply signal measurement, y is the reading obtained from the device with x as the input, and b is an offset error that you may be able to null before the measurement is performed. If m is 1 and b is 0, the output measurement is equal to the input. If m is 1.0001, the error from the ideal is 0.01%.

Parts per million (ppm) is another common unit used to represent accuracy. The following table shows ppm to percent conversions.

Most high-resolution, high-accuracy power supplies describe accuracy as a combination of an offset error and a gain error. These two error terms are added to determine the total accuracy specification for a given measurement. NI power supplies typically specify offset errors with absolute units (for example, mV or μA), while gain errors are specified as a percentage of the reading or the requested value.

Determining Accuracy

The following example illustrates how to calculate the accuracy of a 1 mA current measurement in the 2 mA range of an instrument with an accuracy specification of 0.03% + 0.4 μA:

Accuracy = (0.0003 × 1 mA) + 0.4 μA = 0.7 μA

Therefore, the reading of 1 mA should be within ±0.7 μA of the actual current.