This white paper covers some Frequently Asked Questions we receive from customers about the differences between ISO-17025:2017 vs. ANSI Z540.3.
Q: What are the definitions of ISO-17025:2017 vs. ANSI Z540.3?
A: ISO/IEC-17025:2017 (3rd Edition 2017-11) General requirements for the competence of testing and calibration laboratories ANSI/NCSL Z540.3-2006 Requirements for the Calibration of Measuring Test Equip-ment
Q: What do ISO-17025:2017 and ANSI Z540.3 cover?
A: Both standards address how a Calibration Laboratory should operate.
Q: What is the difference between these two standards?
A: ISO/IEC-17025:2017 essentially covers all the requirement of ANSI/NSCL Z540.3 except for Z540.3 Section 5.3 (b):
5.3 b) “Where calibrations provide for verification that measurement quantities are within specified tolerances, the probability that incorrect acceptance decisions (false accept) will result from calibration tests shall not exceed 2% and shall be documented. Where it is not practicable to estimate this probability, the test uncertainty ratio shall be equal to or greater than 4:1. NOTE: Achieving these requirements may involve adjustment and management of calibration system parameters such as: measurement reliability, calibration intervals, measurement uncertainty, calibration tolerances, and/or guard bands.”
This requires the laboratory to limit the Probability of False Acceptance (PFA) due to the laboratory’s calibration process uncertainty to no more than 2%. Or put another way, the laboratory’s reported value needs to have a 98% Probability of Compliance to the Specification (PCS).
Simplified Definition: (PFA) The probability of making an acceptance decision when the Unit Under Test is observed and reported to be intolerance but is actually out-of-tolerance.
Q: Why is Section 5.3(b) important?
A: No measurement is perfect, there is always some doubt (uncertainty) about every measurement, it’s a gray area. This uncertainty isn’t a problem until you need to call a measurement In-tolerance/Out-of-Tolerance; Good/Bad; Pass/Fail. When the observed value lies close to the test limit, the laboratory’s uncertainty may prevent a definitive answer. There are many ways to handle the calibration process uncertainty when making a statement of compliance. The industry standard practice is to use a Simple Acceptance / Simple Rejection Decision rule, this decision rule allows for a PFA up to 50%. It is important to communicate your desired decision rule to the calibration provider, and/or to know which decision rule the calibration provider will use as a default. Z540.3 Section 5.3(b) defines a maximum allowable PFA of 2% for any decision rule in order to mitigate the risk associate with the calibration process uncertainty.
Q: What can be done is a PFA is greater than 2%?
A: There are several ways to improve the PFA and mitigate risk as identified in Z540.3
Measurement reliability: The measurement reliability of both the M&TE being calibrated and those used as calibration standards affect the estimated PFA. If either or both are increased, the PFA would be expected to decrease. The increase in measurement reliability may be accomplished by shortening calibration intervals, using measurement control processes, etc.
Calibration intervals: Calibration intervals are established to achieve measurement reliability and/or uncertainty goals. Shortening an interval for a given set of M&TE, such as through raising the measurement reliability target, will result in a reduced PFA.
Measurement uncertainty: Calibration process uncertainty. If the calibration process uncertainty is reduced, the PFA is reduced. Reducing the calibration process uncertainty also increases the TUR. The reduction of calibration process uncertainty may be accomplished by use of shortened calibration intervals, using measurement control processes, etc.
Calibration tolerances: If the test tolerances of the M&TE being calibrated are increased, both PFA and TUR are affected. An increase in an M&TE test tolerance would reduce PFA and increase TUR. This improvement may be at the expense of added controls and limiting the application of the M&TE. Guard bands: If a guard band is applied to an M&TE test tolerance to create acceptance limits inside the tolerance, the PFA is reduced. This reduction can be at the expense of an increase in the number of measurement quantities rejected and additional costs for increases in adjustment and repair actions.
Q: What does Transcat do to address this Z540.3 requirement?
A: Transcat has developed methods, policies and procedures to mitigate the calibration process uncertainty impact on the PFA for Accredited calibration by:
- Transcat routinely monitors the reliability of the laboratory standards and adjusts calibration intervals, implements process control charts, uses guard-banding techniques on our standards and strives to improve the calibration uncertainty available to our customers as part of our dedication to process improvement.
- Selecting Laboratory Standards, methods and test points which prevent TUR’s <4:1, thus avoiding a < 2% PFA as per Z540.3 5.3(b) entirely
- Transcat standard practice to identify all TUR< 4:1 and to provide the Calibration Process Uncertainty. This information allows the customer to compute the actual PFA of the reported value if they so desire.
- Transcat also has a service level available that will calculate the Probability of Compliance to the Specification (PCS) for each test point for the customer.
- Transcat can also develop custom datasheet using any specification or tolerance a client chooses to use. The use of custom tolerances can further mitigate or eliminate risks associated with TUR’s <4:1 or high PFA’s.
- Transcat is developing the capability to implement alternate decision rules based upon several of the Guardbanding techniques identified in ANSI Z540.3, ILAC G8 and ASME B22.214.171.124.
Authored by: Phil Mistretta, Manager of Metrology/Deputy Technical Manager Phil Mistretta joined the Transcat management team in 2001. With a background in EMC/EMI compliance testing, lean manufacturing engineering, he has over 30 years of experience in the field of Metrology. He is member of IEEE and ASQ and is an ASQ-Certified Calibration Technician. He is a graduate of Central Texas College and a BS in Engineering Physics at the University of Buffalo.