E for ETALON
15. 04. 2024
The word etalon is first used to refer to a device used to create interference beams of light, essentially consisting of two flat parallel reflecting plates with a fixed distance and (in some types) an adjustable direction. The word probably derives from the Old French word ‘estalon’ or ‘estelon’, which probably originated in Germanic. Interestingly, the French use the word etalonnage for calibration. If we know that simplified calibration is a comparison with an etalon, then this also makes sense. Some other languages also use this form of the word calibration. For example, Serbian also uses etalonage, but it is not in the same language group as French.
Many people outside the profession mispronounce the word as ethanol, but the meaning is of course completely different, even though phonetically it sounds quite similar. Ethanol is alcohol. There is no clever translation for ethanol in Slovene, so the word etalon is used. Dictionary of the Slovene Literary Language says that it is “an exact sample of measure or weight” or “a measure”. Both are true, of course, but the word etalon, according to the International Dictionary of Metrology – VIM (in French: Vocabulaire international de métrologie), means “standard of measurement” or is defined as: “a realisation of the definition of a given quantity with a stated value and associated uncertainty of measurement, used as a reference”.
An etalon can be basically anything that is a reference. The Russian writer Samuil A. Lurie writes “it would be considered the etalon of the architecture in the background, the building to which you give an approving glance as you pass by”. It would be a reference for all other buildings, to put it simply. In a way, the word etalon can be compared to an idol, an ideal. Something that is a reference for you, something or someone that represents an ideal, that serves as a goal to be inspired by. The logic is similar, but not quite the same. An idol can also be something you worship, which is hard to say about an etalon, although in a sense we also worship the etalon, as an ideal. Something that represents a reference value that we all look up to. At the top, or in other words at the top of the metrological pyramid, the very top of the pyramid, are the definitions of units of measurement. There are seven basic units of measurement, which are dealt with or regulated by the aforementioned metric convention, which represents the top of the pyramid. The world convention of accepting units of measurement in practice, the realisation of which we have nowadays reduced to unenviable accuracy. With the approval of the new definitions of units of measurement in 2019, mankind is thus able to set the basic unit of length, the metre [m], with an accuracy of 10-12, the basic unit of mass, the kilogram [kg], with an accuracy of 10-10, and the basic unit of time, the second [s], with an accuracy of 10-15, the base unit for electric current [A], to 10-10, the base unit for thermodynamic temperature Kelvin [K], to 10-5, the base unit for the amount of substance the mole [mol], to 10-7, the base unit for luminous intensity candela [cd], to 10-5.
We can probably do better than that, but the definitions are about as accurate as written at the moment, and as such are sufficient for the state of the art. The definitions of basic units of measurement, like other technological developments of mankind, must be adapted and fulfilled at a similar pace to meet the needs that developments have. After all, if the end-user of a measuring instrument, say in the automotive industry, wants to measure something to an accuracy of 10-6 [μm] (micrometre; 0.0000001 m or 0.001 mm), which is not so unusual any more, and which practically happens every day, at all levels, with a perfectly ordinary measure called the Micrometers (which has existed for quite a few centuries; the first micrometre is said to have been made by James Watt in 1776 – by the way, this is a year before the Patent of Cimetry, the first law in the field of metrology on Slovenian soil and in the wider world), the definition of the basic unit of measurement, the metre, has to be much better. This enables so-called metrological traceability, which is a continuous chain of comparisons. Thus, calibration at the lowest level (micrometre calibration) can be obtained by the user almost around every corner at a reasonable cost. If that same micrometre had to be calibrated directly to the definition, it would cost considerably more and even calibration would not be available everywhere. Metrological traceability in this case goes something like this. The micrometre is calibrated with gauge blocks (accuracy somewhere in the range of 10-7), the gauge blocks are further calibrated by mechanical or optical comparison (accuracy of 10-8 or 10-9), then comes the laser interferometer (accuracy of 10-10), and finally we come to the basic unit of measurement. Since the number of comparisons (calibrations), and hence the cost, is spread over several gauges, the whole is more accessible and less expensive. The system works and delivers the expected results.
Returning to the etalon. It has to meet something other than the approval of the writer. It must have an exact known value, which is established by comparisons as described in the previous paragraph, it must guarantee this accuracy at all times, at least within some known or predictable band, which we normally call long-term stability, it must have an appropriate form to enable downward comparisons to be made and to yield sufficient accuracy, and it must allow for general use. Probably the most familiar etalon to the masses is the weight, which also historically has a long tradition. Already a litre of water is a well-known domestic standard for 1 kilogram and is a reasonable approximation for most things outside the discipline. The first definition (admittedly still outside the Metric Convention) was set using a litre of water. Today, most of the world (at least in the scientific and professional world) uses weights in the form of the basic unit of measurement kilogram, and all its derivatives (prefixes). The most widely used recommendation (standard) for weights in the world is OIML R111-1, which defines weights from 1 milligram up to 5 tonnes. Of course, there are also weights less than 1 [mg] and greater than 5000 [kg], but these are standardised. The material (steel is the most commonly used, but aluminium or brass can also be used), the dimensions (in principle, milligram weights are wire or lamina, while gram and kilogram weights are cylindrical), the surface finish (polished or coated), and the accuracy – or rather the accuracy class – or the maximum permissible error of a particular weight with respect to the accuracy class, are all prescribed. And today, a litre of water is determined by a kilogram, not the other way round, because gravimetry (mass) is much more accurate than volumetry (volume). Practically all volume measurements at the highest level are made by mass.
Etalons come in a wide variety of shapes and designs. In principle, anyone can produce an etalon. After all, a worker in the forest who cuts down and saws trees most often makes his own (1 metre) standard so that he can saw all the logs to the same length. This can also be called an etalon. Its accuracy is, of course, considerably less, but the point of an etalon is not only its maximum accuracy, it is above all its usefulness and meeting the needs at a given moment. The essential point in all this is that the user of the etalon is aware of the etalon’s ability not to overestimate its accuracy and only uses it for its intended purpose. After all, quality is, in simple terms, meeting the requirements and expectations of the user (customer). There are many more examples of this kind. A soldier can measure (estimate) distances quickly and efficiently using body parts. In a given situation, probably quite well enough. We humans have a very good sense of the dimensions we can grasp. In the millimetre range, up to a metre or even more, most people will be able to estimate length quite well. Shorter or longer distances, however, present us with greater difficulties, and the differences are much greater. Similarly with temperature. We can feel the differences very quickly at so-called room (living) temperatures. We can quickly feel a difference between 20 °C and 21 °C, whereas at -10 °C and -11 °C, let alone at lower or higher temperatures, we do not feel any difference.
That is why man has developed criteria. Because we are not objective, or we are not able to detect differences. Measures provide objective evidence for a characteristic that we measure. And standards are a kind of means of measurement that provide stability, that allow us to derive comparisons, that a metre, a kilogram, a degree Celsius, etc. is the same everywhere in the world. At least approximately the same. Good enough, I would say. That is, after all, the basic principle of metrology. Good enough. Measure well enough. At a given time, in a given situation, according to requirements and needs, satisfactory. And etalons are one of the pebbles in the mosaic. They are one of the pillars of metrology in the world. The better the etalons, the better the metrology. They are still managed by man, but the etalon is the means where the human factor is eliminated as much as possible. In the letter Č I was thinking of man, and man has no limits. In a positive or negative direction, so that we understand each other. We have learnt from history that even in the field of etalons there have been mistakes, serious mistakes. After all, the first definition of the metre, which was set by means of the so-called H bars, was a grave mistake. It was not fatal, but it was subsequently discovered (several decades later) that the H bars were about 0.2 mm too short. Not because someone made them too short, but because of a calculation error which did not consider that the earth is a sphere. How interesting that we can prove that the earth is round even through units of measurement. Yet the “flat earth” theory has followers all around the globe (all AROUND the GLOBE). They even wrote it themselves. Well, there you have the answer to how good the etalons are. As much as the people who operate them. Which is true of everything around us, and etalons are still a pretty safe thing as far as I’m concerned.
Primož
Next time, 1st of May 2024, SHAPE (FORM)