Calibration of Test and Measurement Equipment - Technology Overview
The International Office of Weights and Standards defines calibration1 as a set of operations that establish, under specified conditions, the relationship between values of quantities indicated by a measuring instrument and the corresponding value realized by standards-in other words, comparing the instrument to a reference standard. As a result, the calibration lab may decide to adjust the instrument as per calibration results to bring it closer to the reference standard.
In most cases, calibration is performed by comparing the instrument to be calibrated to a reference instrument (like the IQS-1500 Calibration Power Meter) called the standard, as it has been calibrated by a national calibration laboratory like NIST or NPL. Other instrument types can be verified by using a natural physical constant; for example, a wavemeter that is verified using a laser locked onto an absorption line of acetylene (e.g., P13 C2H2).
Standards and Procedures
When calibrating instruments, laboratories will base their calibration procedures on calibration standards or guidelines set by international standard committees like the International Electrotechnical Commission (IEC) or Telecommunication Industry Association (TIA). These committees, in which EXFO is actively involved, will review the critical parameters to test for each type of instrument and will write procedures that will ensure that the test instrument is properly calibrated.
ISO/IEC 17025
ISO/IEC 17025 was introduced in 1999 and updated in 2005 by the International Organization for Standardization (ISO) as well as the IEC, and it is the most important metrology standard for setting requirements for calibration laboratories. It applies to all organizations performing tests and calibrations and includes both management and technical requirements.
The ISO/IEC 17025 standard covers issues like test and calibration methods, personnel training, measurement traceability, uncertainty analysis, calibration certificates and test reports, and many other relevant issues.
Uncertainty
Every measurement
has an uncertainty. Therefore, when testing or calibrating test
instruments, it is very important to take the measurement uncertainty
into account. Lets look at an example, in which one wishes to
ensure that the output power of a laser meets the minimum output power
requirement:
- Specification: laser’s
minimum output power ≥
–3 dBm
- Uncertainty (U):
0.3 dB
- Measured Output power:
-2.8 dBm
The quick conclusion after this test would be that the instrument complies with the manufacturers specification since the output power is greater than -3 dBm. Nevertheless, this would only be true when measurement uncertainties are not taken into account. But when we wish to guarantee with a high level of confidence (= 95 %) that the output power of the source is = 3 dBm, then measurement uncertainties must be taken into account:
- Tolerance for a level of confidence of 95%: Pout ≥ Spec
+ 0.825 * U:
≥(–3 + 0.825* 0.3) dBm
or ≥ –2.75 dBm
- Tolerance for a level of
confidence of 97.5%: Pout
≥ Spec + U
≥ (–3 + 0.3) dBm
or ≥ –2.7 dBm
Therefore, a measured output power of 2.8 dBm means that it is possible that the source output power is below the manufacturers specification of 3 dBm.
This way of judging a units compliance to specification by taking into account measurements evaluating the test results of tests is one of the key points of the ISO/IEC 17025 standard. Compliance to specifications should always be specified by taking into account the uncertainty, not only by a simple Pass/Fail statement.
The quick conclusion after this test would be that the instrument complies with the manufacturers specification since the output power is greater than -3 dBm. Nevertheless, this would only be true when measurement uncertainties are not taken into account. But when we wish to guarantee with a high level of confidence (= 95%) that the output power of the source is = 3 dBm, then measurement uncertainties must be taken into account:
Table 1- Four zones of compliance, for a level of confidence of 95%
By looking at the graph above, we can see that when uncertainty is not taken into account, the instrument is compliant with the manufacturers specifications. However, in the top part of the graph, we see that if we take the 0.3 dB uncertainty into account, the unit is Within specification*, which means that there is a risk that the unit may not be compliant with specs as the output power can be between 2.55 dBm and 3.05 dBm.
The end user then has to decide if it is acceptable for him, based on applications and requirements, to use the instrument in such condition or if another verification or repair is needed.
Traceability
Another key point of the ISO/IEC 17025 standard and good calibration procedures is the traceability of the measurements.
The International Vocabulary of Basic and General Terms in Metrology defines traceability as:
"property of
the result of a measurement or the value of a standard whereby it can be
related to stated references, usually national or international
standards, through an unbroken chain of comparisons all having stated
uncertainties."
For example, when calibrating a customer`s power meter, its possible to use an IQS-1500 Power Meter calibrated by EXFO. When doing so, EXFO uses an IQS-1500 which has been calibrated by the NIST. Therefore, measurements performed on the customers unit are traceable to NIST standards.
References
1. BIPM, IEC, IFCC, ISO, IUPAC, IUPAP, OIML: International Vocabulary of Basic and General Terms in Metrology (VIM), 1993.