Calibration Quality Assurance

Download as pdf or txt
Download as pdf or txt
You are on page 1of 1

eer’s

Engin ook
N oteb
Calibration —
Quality Assurance

Reading Accuracy Specifications

Whether you are evaluating process instruments or calibrators, Selecting Standards of Appropriate Accuracy
one of the biggest challenges in calibration management is
Many organizations use the rule of thumb that calibration
reading and comparing accuracy specifications. Each
standards have an accuracy ratio of 4 to 1 between the
manufacturer has a different expression for accuracy, making it
standard and the Unit Under Test (UUT). Stated another way,
Calibration

difficult to evaluate and compare devices.


the uncertainty in your calibrator or standard should be no
An accuracy specification is easier to comprehend if you greater than 1⁄4 the specified uncertainty in your UUT. For
think in terms of the uncertainty in the measurement. In a example, if a gauge has an uncertainty of ±1 PSI at its full
simple example, a 100 PSI test gauge has an accuracy scale reading of 100 PSI, you would want to select a
specification of ±0.25% of full scale. The uncertainty of any standard whose uncertainty was no greater than ±0.25 PSI
measurement made with this gauge can be calculated by at 100 PSI.
taking the full scale range (100 PSI) times .25% (.0025).
Comparing accuracy is not always simple because there are
Therefore a measurement of 30 PSI made with this gauge has
many different ways accuracy can be expressed, and ranges
an uncertainty of ±0.25 PSI or is between 29.75 and
on standards usually do not match ranges on UUTs.
30.25 PSI.
In selecting standards, first choose one whose range is as
Accuracy/uncertainty specs may be stated a number of
close as possible to that of the UUT, then compare accuracy
different ways: % of Scale, full scale(FS), or range, % of
by applying the specification math to key calibration points.
reading, or of indicated value. In addition, on digital
instruments, there will typically be an additional specification
of ± a number of counts or digits. If you see a specification
For example:
expressed in PPM, 1 ppm is .0001%, so 15 ppm is 0.0015%.
• The UUT is a 30 PSI Transmitter, ±1%
To interpret the true capabilities of a device, you need
The proposed standard is a 100 PSI test gauge, ±0.25%
to apply the math in an accuracy specification to a
measurement you would typically make. • UUT uncertainty is 30 PSI x 1% or ±0.3 PSI
Standard uncertainty is 100 PSI x 0.25% or ±0.25 PSI
• Ratio is 0.3 PSI ÷ 0.25 PSI or 1.2 to 1.
Two examples: This would not be an appropriate
1. A mA calibrator has a range of 24.00 mA, accuracy is 0.1% standard for this calibration.
of FS ±1 digit. Such calculations can be time consuming, and they
certainly are not something you want to do every time you
• Full Scale is 24.00 mA
do a calibration. Ideally, the determination should be done
• ± digit specs are always farthest to the right of the once, and the appropriate standard should be written into the
decimal, in this case,+0.01 mA calibration procedure. This practice, coupled with sound
• At 20.00 mA this instrument is ±(24.00 x 0.I%) + 0.01 management, will help insure the validity of your calibrations.
• After rounding, the uncertainty at 20.00 is ±0.03 mA
2. A pressure transmitter has a range of
0–30 PSI, a 4–20 mA output, and an overall accuracy spec
of 0.5% of full scale.
• Full Scale is 30 PSI
• Full Scale output is 20 mA
• Uncertainty in mA at 30 PSI is 20 mA x 0.5% or ±0.1 mA
If you want to do a comparison of two or more different
devices, apply the math in the specifications to the same
measurement value (for example 20 mA) made with the
various instruments. This will give you a realistic picture of how
instruments compare.

You might also like