Is there a standard for temperature similar to those for length, mass, etc.?

Factual Questions
21

Former temperature metrologist here. :slight_smile:

As noted by a couple others, there are two defined temperatures:

  1. Absolute Zero (defined as 0 K)

  2. The triple point of VSMOW water (defined as 273.16 K = 0.01 °C)

Again, these are defined; they will never change. All other temperatures are based on these two standards.

Interesting tidbit of info: it is impossible to measure the temperature of the triple of water. It is defined, remember? You can’t measure something that’s defined. It is used to measure the temperature other things.

The triple point of VSMOW water is pretty easy to achieve in an analytical lab. All you need is a water triple point cell (around $1000) and some dry ice. After forming the mantle, it will be at 0.01 °C for two or three hours, and then it will melt. (Well, not quite. A practical water triple point is not exactly 0.01 °C. You need to use correction factors.) During that time, you can stick any thermometer into the well for calibration. Once it melts, you simply have to reform the mantle.

But also as noted by others, there are other fixed point temperatures that can be used for calibration. These include the triple point of mercury (−38.8344 °C), the melting point of gallium (29.7646 °C), and the freezing point of indium (156.5985 °C). But these are not defined temperatures; they are simply known out to around ± 0.00003 °C or so. In the future they could change. As an example, we might later discover the freezing point of indium is closer to 156.5984 °C than 156.5985 °C. Or the freezing point of silver is closer to 961.79 °C instead of 961.78 °C. But absolute zero and the triple point of VSMOW water will never change. They’re (Again!) defined, and are used as a basis for determining the temperature of all other things.

So… how does NIST know what the temperature is for the triple point of mercury? Or the freezing point of zinc? Or the freezing point of indium? They use gas thermometers. You can’t buy these; they are very special, very difficult to use, and very expensive. Physicists like them because they’re based on physical laws. They first taking a reading w/ a gas thermometer in the triple point of water, and then they use it to measure the temperatures of the other fixed points.

Note that the melting point of water and the boiling point of water are no longer considered fixed points. And because of that, it is technically incorrect to say the melting point of water is exactly 0 °C, or the boiling point of water is exactly 100 °C. The latest measurements suggest the melting point of VSMOW water is around 0.000089 °C. So it’s good enough for many applications, but temperature metrologists still don’t like it. :slight_smile:

So after NIST measures the temperature of all of its fixed points using fancy gas thermometers, all the fixed points (including the triple point of water) are available for calibrating practical standard thermometers. The most common being standard platinum resistance thermometers (SPRTs) and type S thermocouples.

Thousands of calibration labs across the world use an SPRT and type S thermocouple as their standard interpolation instruments for temperature calibrations. (Most also have a triple point of water cell. The cell is necessary when using an SPRT, as you must first stick the SPRT in the cell to show the SPRT what 0.01 °C looks like, and then you use it to calibrate other sensors in a temperature controlled bath.) At these labs, they use the SPRT and type S thermocouple to calibrate a wide variety of temperature sensors and readouts using a comparison method. These labs will periodically send their SPRTs and type S thermocouples to NIST (or another qualified lab) for calibration. We had to send our SPRT back to NIST whenever its resistance at the triple point of water drifted by more than 0.001 Ω.

22

You buy a TPW (Triple Point of Water) cell and a special temperature controller to put it in, and chill it until there’s a mantle of ice floating around its central thermometer well. That’s the big standard. There are other ones with indium, tin, aluminum, and so forth.

23

We had a YSI water triple point maintenance bath. Was really nice, as it allowed a triple point water cell to last for weeks instead of hours.

Most run-of-the-mill cal labs simply have an SPRT and one or two water triple point cells. Other cells such as indium, tin, etc. are only used by primary standards labs. These fixed point cells are used to calibrate SPRTs for the run-of-the-mill cal labs.

24

I have posted this before, but I see that I need to repeat myself.

The foot is approximately the length of an adult human foot. Everybody can visualize how long a foot is through personal experience.

English units of length are much more sensible in other ways, too. In metric, when most people give their height, they have to say 1.xx meters, or else in the hundreds of centimeters. The foot is a very reasonable intermediate measurement.

That is just one of the great many examples of how English measurement is so much more sensible and practical than metric.

25

Yes, six feet is a normal height, but two meters is almost grotesquely tall, no? Similarly, a 2x4 plank is precisely 1½" by 3½" — it would be cumbersome to ask for 38.1 mm by 88.9 mm.

As for weight, however, I must disagree. 14 stones is a nice round number (if you’ll pardon the pun) and 14.3 stones can be rounded to 14 without complaint. Unfortunately, Americans haven’t used stones since the Revolution AFAIK, and we’re stuck with pounds. That’s why I prefer kilograms. Some days I weigh in at 90 kg, some days 91. Same-same right? But these numbers are absurdly different (198 vs 201) when you convert them to the evil American pound. :stuck_out_tongue:

26

I disagree. The foot is nice, yes. Metric would be ever so slightly better if it had a unit about that long. Where I disagree is your last sentence; the foot is about the only thing I miss. All the rest of everyday-use metric is either noticeably easier and better than the old system, or it makes no difference. Celsius degrees are better and easier. Metric paper sizes are better and easier (wish we used them here). Metric weights in the grocery store - half a kilogram is about a pound, anybody stuck in pounds can easily handle that.

27

O Anglophiles, why the need to puff up English units of measurement as opposed to, say, Russian ones? Anyway all this anthropocentric metrology was already commonplace stuff by the time of Ancient Mesopotamia and Egypt. If you think about it, though, you have to at least admire the audacity of the revolutionary French heretics in booting Man right off the stage like the insignificant nonentity he is, and replacing the marks of his dirty limbs by reflections of the pure heavenly spheres themselves.

28

Don’t fool yourself: The only non-anthropocentric measurement system is the one where c = h = 1, the natural units, not the metric system or the SI or whatever it’s called now.

In the natural units, all intervals, timelike, spacelike, and null, are measured in the same unit. In less technical terms, in the natural units time and space are measured using the same unit.

It’s as revolutionary as measuring heights and widths using the same unit, only moreso, and, therefore, only physicists use it.

29

How about this time you take some time actually thinking about the replies?

Feet vary in size, anyone growing up with metric can visualize a meter to a similar level of accuracy. (In my case, very low accuracy, but that applies to all units of measurement.)

I’ve never met someone who didn’t include inches when giving their height in English, and fail to see how “One eightysix” is inferior to “Six foot one”.

30

Another sensibility to the USA units is they are slightly smaller than either Imperial or metric … an American can put his product on the world market for a bit lower price per ton (2000 lbs) than a European “ton” (2200 lbs) yet still be charging slightly more per pound [ka’ching] … same can be said of USA gallons vs. Imperial gallons and USA yards vs. the meter …

Most European countries use a decimal unit of currency and metric units making it very easy for the consumer to calculate in their head what the best value is … whereas the United States uses a decimal currency and fractional units making it very difficult to find the best value even with a digital calculator … much easier to rip off the consumer here in the good ol’ USA …

Put one box of product at 1 lbs 14 oz and another at 3 lbs 2 oz at twice the price … the typical harried shopper will think “triple the product for only twice the price” [ka’ching] …

Nothing wrong with a measurement standard hellbent on increasing profits …

31

My feet are a ladies’ 36 EU.There’s a guy in this same factory who wears a mens’ 51 EU.

According to this nifty chart, there’s a difference of 10.2cm, which roughly comes up to 4 inches, or 1/3 of a foot.

Approximate indeed. With that level of accuracy, pi equals 2, or maybe 4.26896764366596.

32

How hard is it to visualize a meter, anyway? I mean, even us English-unit-only folks have seen a yard stick or been to an American football game. Or just take a slightly longer than average step and, voila, you’ve got a meter. It’s not any less intuitive than a foot, IMHO. When I think of, say, the distance from one side of the road to the other, or the passageway between my house and the house next to it, etc., I think in terms of yards or meters. I find that a lot easier to visualize and estimate than using something like feet. So it also kind of depends on what you’re measuring, but even though I grew up with English units, if I had grown up with metric, it would have been as intuitive, plus easier to work with since I don’t have to memorize conversions between different sized units.

33

It does: 30 cm.

12 inches is 305 mm, or thereabouts, 30 cm is close enough to “about a foot long” for most purposes. Or you could say “305 mm” and be more exact, if that’s needed. Anyone familiar with metric measurements will know how long 30 cm is without having to resort to idiocy like “as long as an adult foot”.

34

The only fault of the metric system is they should have gone base-12. I mean, we should have revamped all math into base-12 including metric.

And what about metric time? Instead of hours, days, weeks and years, we need kiloseconds, megaseconds and gigaseconds.

35

The average length of a month depends on whether the year is divisible by 4 but not 400 … yeesh …

36

What’s so special about an adult foot? Anybody can visualize anything if that’s what their personal experience is. If you are from planet Sartokl then then everybody knows that 1 aidjez is approximately the length of the tentacle of a gremtl just out of the larval stage, and the English system makes no sense at all. And everybody who grew up with the metric system can visualize a meter quite readily.

But almost nothing occurs naturally as an exact multiple of feet, so this argument falls apart for anything other than things like boards that are arbitrarily specified to be 8 feet. When I give my height I say 5 feet 9-3/8 inches, having to combine two different units that have an arbitrary relationship. How is that better than 1.76 meters?

Repeating something doesn’t compel it to be right.

37

Time is tricky because the solar year is not a whole-number multiple of days, but we want our calendar year to have a whole-number multiple of days. We could have 10 months every year that are each exactly 36.52421875 days long but there is something convenient about the solar day that we like to keep.

38

I agree. Having more easily-usable divisions between (some of) the units is about the only advantage Imperial and US Customary have over metric.

39

If some American/old English units have convenient conversion factors, it’s by pure coincidence. Yeah, a power of two for ounces in a pound and an abundant number for inches in a foot can both be convenient, but what the heck is the logic behind 14 pounds in a stone, or 231 cubic inches in a gallon?

40

Something I wrote before as well.

I grew up during the change from Imperial to Metric in Oz. My school year was the event horizon for the introduction of metric in the curriculum, if you were a year older than me you went through with imperial units. I am perfectly comfortable with both imperial and metric for most simple tasks. Feet and inches, meters, makes no difference to me. Ounces and pounds, grams and kilograms, pints and gallons, versus litres, much the same. There is absolutely nothing intrinsic to the units that makes them somehow more natural or intuitive.
There are odd hangovers from the imperial era. I buy wood in pre-cut multiples of 300mm. A sheet of MDF is a 1.8x1.2. A 6x4 in old money. 2x4" lumber was never that size (being the rough sawn dimensions, not the dressed dimensions) and I now buy the metric equivalent.
When it comes to actually designing something nobody sane would go for imperial. Not unless you went for decimal feet. Which gets just plain silly. I design things in metric. The only time I see imperial units is in drill sizes - which is annoying, but not difficult to manage - I have sets of both imperial and metric drills. (So far avoiding the need for a numbered drill set.)
For instance I just finished building a pair of bookshelf speakers. All the design is in metric. It makes the design process workable. Things like the internal volume of the boxes is critical, I can work in litres trivially. Working in cubic feet is dire at best, and the important calculations of the physics of the system are metric from the driver’s parameters right through to the dimensions I set the table saw to when cutting wood. OTOH, I can tell you exactly how big the thing is in imperial units, simply because I can handle both. But that is just a historical curiosity.