Understanding the performance of test and measurement equipment

Assumptions about an instrument can lead to unpleasant surprises during the calibration cycle.

Test and measurement equipment play a critical role in most production and development processes. These instruments are used to make decisions regarding the effectiveness of processes and to ensure product quality. It is no surprise, therefore, that the need for regular calibration of test and measurement equipment is included as part of most major quality standards and is a commonly reviewed area for auditors, whether they are from an internal department, accreditation body or a regulator such as the FDA. Most quality systems will also define the actions that must be followed in the case of an instrument being found out of specification during a calibration. What sometimes does not get enough attention is the selection of test equipment to reduce the risk of an out-of-tolerance condition affecting the process.

There is a fine balance to be considered when selecting equipment. Cost is always an important consideration but let us look at the other critical issue: appropriateness of the instrument to the process. It is critical that the instrument selected can meet the measurement requirements of the process not only on the day it is delivered new, but also for the time period between calibrations.

All measurement instruments are prone to drift over time. Unfortunately, from a process view this means there is always some risk that instruments currently in use are actually out of specification. To mitigate this risk, calibration intervals are assigned based on manufacturer specifications, process requirements, equipment history and acceptable risk requirements.

In most cases the manufacturer’s specifications are weighted heaviest when determining suitability of equipment for a process. It can be difficult to make a proper determination of an instrument’s expected long-term performance because all of its specifications must be looked at to gain proper insight into its performance. It is not sufficient to look solely at the instrument’s accuracy. That usually does not take into account long-term (or one-year) drift of the instrument, or may only be applicable to a narrow operating environment. Not properly understanding the performance of an instrument will lead to unpleasant surprises during the next calibration cycle.

Some instrument types and applications tend to be easier to define than others. A high-end digital multimeter used in a laboratory setting can be relatively straightforward. Many manufacturers will state 24-hour, one-week, one-month and one-year accuracies as well as provide guidance about the effect of temperature on the readings. The behaviour of these types of equipment is well known and, in the grand scheme of metrology, relatively predictable. On the other end of the spectrum are instruments such as relative humidity meters. These are used in such a wide variety of operating conditions and environments that it is difficult for manufacturers to specify the right instruments for every application. Every environment is different and could have a different effect on the sensors.

When looking at specifications, always check to see if all the factors are included in the stated accuracy. Many times, the stated accuracy is only valid for a specific temperature range, or it may not include the calibration uncertainty, or perhaps the long-term drift is not included. It is important to take a careful look at what is and is not being included in an accuracy specification. If it does not explicitly say that long-term drift is included in the accuracy, it likely is not and must be taken into account separately.

The end user needs to be aware of requirements of the process and how the reference equipment will support them. It is critical to gather all the information regarding an instrument before approving it for a process. The repercussions of not knowing are typically far more expensive that the initial effort.

By Michael Boetzkes, the quality manager of life sciences for Vaisala Canada Inc’s Life Science and High Technology Division. His background in physics, Boetzkes joined the Veriteq company in 1998 and served as calibration lab manager and production manager. Then, as vice president of quality, he developed the quality systems to meet the requirements of life science and regulated customers before Veriteq was acquired by Vaisala in 2010. Boetzkes has furthered his expertise in calibration and quality systems to include relative humidity.

This article first appeared in Quality Digest Daily (www.qualitydigest.com).