We all know stories (many times contentious) about how units of measurement were derived, such as a foot being equal to the length of King Henry I’s foot; the inch being the width of the thumb; the cubit being the distance from the elbow to a line between the thumb and another finger; and degrees of arc being a close match to the number of days in a year and easily divisible by every number from 1 to 10 (except 7).

The point is that systems of measurement are associated to standards or to events that can be referenced or repeated. The reference becomes the “gold standard” against which everything else is measured. Early standards were quite crude by today’s values, and, as time went by, tools and skills improved and more constant, reliable scales became necessary. In the 21st century, many of the standards for units of measurement are precise, sometimes to the atomic level. One U.S. foot equals 1,200/3,937 meters. One second of time equals “the duration of 9,192,631,770 periods of the radiation corresponding to the transition between the two hyperfine levels of the ground state of the cesium-133 atom.”

There are recognized systems of measurement that relate to standards. A direct measurement uses a scale that directly correlates to a standard. If I say the part is exactly 1.0435-in. long, someone could make a part that would exactly mate with it, even before that person ever sees an actual part, because of the common scale.

Ah, if only it were that simple!
It’s important to understand there are two uses of standards in the context of measurement. One is the reference standard for specific units of measurement. In the earlier example of the unit of time, the reference standard for one second is the behavior of a cesium-133 atom. The purpose of this standard is to have a recognized method for verifying (calibrating) any devices that measure in seconds, multiples and fractions of seconds. It’s not very likely that anyone besides a standards organization such as the National Institute for Standards and Technology would actually possess such a reference material or the means to monitor its behavior to match the definition. Yet, having such an agreed-upon standard for a second means that seconds can be measured directly by a multitude of devices.

In the case of length units such as the inch, there are other recognized standards that can be used for verification purposes. Standards organizations have the means to verify the units. This brings up the other context for standards—physical standards based on those units of measurement.

A common example is the gage block. Gage blocks are available in a range of lengths and are used for comparative measurements and for calibrating devices that measure length. Other physical standards for dimensional metrology include scales and reticles such as those used for verifying linearity of translation stage travel or image quality of optical systems.

These physical standards are based on the references retained at the standards organizations. You may be familiar with the term “NIST-traceable.” This means that a physical standard such as a gage block or a reticle has one or more dimensions that were calibrated by NIST, or by an intermediate calibration lab. The deviation of the item from the standard is noted so it can be used as an in-house, working reference. It’s a proxy for the actual reference standard that companies can use directly. In King Henry’s time, this would be similar to putting his foot on a piece of metal and cutting the piece to the exact length of his foot. That metal piece would be a physical reference to, or a proxy for, the standard, his foot. The metal piece could be considered “traceable,” because its length can be shown to be directly traceable back to the standard of measurement, the king’s foot.

Using reference standards
It’s important to point out that just because a gage block can be NIST-traceable it doesn’t mean that all gage blocks are traceable to a standards organization.

With a reference standard on hand, a company can calibrate its measuring devices, verifying the accuracy of their measurements. However, just having a reference standard doesn’t mean highly accurate measurements are a given. A number of factors come into play. One is environment.

When you look at a traceable standard, the certificate that accompanies it includes detail about the environmental conditions under which the item was measured (calibrated). Implicit in that certificate is that all bets are off regarding the reported value if environmental conditions are not met. For example, gage blocks are usually "soaked" to a granite plate so they can reach a particular thermal equilibrium within a stable environment. Simply holding such a block in your hand can make it longer than the reported value due to thermal expansion from heat.

Proper care and periodic recertification of standards are requirements.

Absolutely!
There’s a lot more to measurement than most people think. Scales of measurement for direct and comparative measurements are important. Reference standards help ensure that the performance of individual measuring devices can be verified for accuracy. They make it possible to improve measurement accuracy, but simply having standards available doesn’t guarantee accuracy. All measurements relate to something. Measuring devices and systems are simply the means for presenting those relationships.