Talk:Kelvin

Is the Kelvin temperature scale actually supposed to be written as a lower-case k? I thought k stood for kilo- and K stood for Kelvins. --24.72.34.179 22:44, 14 Jun 2004 (UTC)

I think they meant that:

1. A "kelvin" is written with a lowercase "k", just like "ampere" is written with a lowercase "a", even though it's named after a person too.
2. However, the abbreviation for a kelvin is an uppercase "K", just like the abbreviation for an ampere is an uppercase "A".

200 K = 200 kelvins
200 A = 200 amperes

Perhaps the wording needs some clarification, or perhaps it's unnecessary. -- Curps 23:46, 14 Jun 2004 (UTC)

Page 31 of the United States National Institue of Standards and Technology (NIST) Special Puplication 811 (Guide for the Use of the International System of Units (SI)) states that unit names should be spelled out in full like any other English noun, with lower-case letters, except at the start of a sentence or in a title. However, if you write "degree Kelvin", Kelvin should be capitalized (since it's a type of degree, same way you'd write "Kennedy High School" and "high school") The publication actually uses celsius for the example but it's the same principle. A capital K should be used when abbreviating, since the unit is named after a person. -- Uberdog 02:14, 30 Jun 2004 (UTC)

According to the BIPM (see here) (http://www1.bipm.org/en/si/base_units/kelvin.html), the "degree Kelvin" ceased to exist 36 years ago:

The 13th CGPM (1967-1968, Resolution 3) adopted the name kelvin (symbol K) instead of "degree Kelvin" (symbol °K) and defined the unit of thermodynamic temperature as follows (Resolution 4): ...

On the other hand, Celsius has a capital C because it hasn't been adopted by the SI. -- Heron 08:25, 30 Jun 2004 (UTC)

No. Celsius has a capital C because it's one of the quirks of the English language (these are language-specific rules; it's different in German, for example, with all nouns capitalized) that the nouns naming units after people are not capitalized, but the proper adjectives identifying particular ones of ambiguous unit names are capitalized. Thus, when kelvins were called "degrees Kelvin" they also had a capital K, and the R is capitalized in degrees Rankine.

Furthermore, you are also wrong about degrees Celsius not being adopted by the CGPM. They are indeed part of the SI (which, as a system of measurement, cannot "adopt" anything), a derived unit with a special name.Gene Nygaard 13:39, 10 Dec 2004 (UTC)

In case anyone's interested, here's a NASA styleguide (http://history.nasa.gov/styleguide.html), which is where I confirmed my suspicion that a space belongs between the number and the symbol for Kelvins, unlike in many places in Wikipedia. NASA occasionally has trouble with units, but hopefully they can spell them right. -- Tantalate 01:47, 4 Aug 2004 (UTC)

I agree. The American NIST checklist (http://physics.nist.gov/cuu/Units/checklist.html) mandates a space between every number and its unit. The official SI brochure always uses a space (e.g. here) (http://www1.bipm.org/en/si/si_brochure/chapter5/5-4.html), but I can't find an explicit statement on the BIPM site that a space is required. -- Heron 08:13, 4 Aug 2004 (UTC)

This topic comes up occasionally in other discussions. ISO 31-0 and the UK equivalent BS 5775 apparently recommend a space before the unit, but I have not seen the original text of either. I also find the recommendation in:

• US NIST publication 811 (http://physics.nist.gov/Pubs/SP811/sec07.html)
• US NIST checklist (http://physics.nist.gov/cuu/Units/checklist.html)
• UK NPL (http://www.npl.co.uk/reference/si_conventions.html)
• IEE guide (http://www.iee.org/EduCareers/ProfDev/Guides/UNITSANDSYMBOLSFORELECTRICAL.doc)
• misc.metric-system faq (http://www.cl.cam.ac.uk/~mgk25/metric-system-faq.txt)
  

The scope of the references certainly includes metric units. Whether the authors would recommend the same format for non-metric units is unclear to me. It may be implicit in the IEE reference and the unseen ISO and BS references. I would be surprised if any respectable author/editor would specifically want all copy to have spaces before metric units but not before imperial units. In summary, there are respectable references that say a space should be used and none that say a space should absent. Where each character is costly (such as on a cellular phone display), I don't think people worry too much if the space is absent, but Wikipedia pages have plenty of room for space characters.

Bobblewik 09:17, 7 Aug 2004 (UTC)

Any reason why both 273.16 ("...one kelvin is the fraction 1/273.16 of...") and 273.15 ("°C = K − 273.15") is used? which is the right one?

They are both right. 273.16 is the definition of the kelvin with respect to the triple point of water. 273.15 is the offset between the Celsius scale and the kelvin. See the BIPM website (http://www1.bipm.org/en/si/si_brochure/chapter2/2-1/2-1-1/kelvin.html). --Heron 14:18, 19 Oct 2004 (UTC)

Definition and comparison with Celsius

I just had an edit of mine reverted and I'd like to explain myself. The BIPM brochure on kelvin (http://www1.bipm.org/en/si/si_brochure/chapter2/2-1/2-1-1/kelvin.html), which link I added to the article, states the following (emphasis is in the original):

The 13th CGPM (1967-1968, Resolution 3) adopted the name kelvin (symbol K) instead of "degree Kelvin" (symbol °K) and defined the unit of thermodynamic temperature as follows (Resolution 4):

The kelvin, unit of thermodynamic temperature, is the fraction 1/273.16 of the thermodynamic temperature of the triple point of water.

Because of the way temperature scales used to be defined, it remains common practice to express a thermodynamic temperature, symbol T, in terms of its difference from the reference temperature T₀ = 273.15 K, the ice point. This temperature difference is called the Celsius temperature, symbol t, and is defined by the quantity equation

<math>t = T - T_0<math>

I re-stated and summarized this as follows:

The kelvin (symbol: K) is the SI unit of temperature, and is one of the seven SI base units. It is defined as the fraction 1/273.16 of the thermodynamic temperature of the triple point of water. A temperature given in kelvins, without further qualification, is measured with respect to absolute zero, where molecular motion stops. It is also common to give a temperature (a so-called Celsius temperature) relative to the reference temperature of 273.15 K, approximately the melting point of water under ordinary conditions.

I think this is a fairly faithful rendering of the BIPM's description, but I'm open to discussion on the point. Our prior (and, after the revert, our current) definition reads as follows:

The kelvin (symbol: K) is the SI unit of temperature, and is one of the seven SI base units. It is defined by two facts: zero kelvins is absolute zero (when molecular motion stops), and one kelvin is the fraction 1/273.16 of the thermodynamic temperature of the triple point of water. The Celsius temperature scale is now defined in terms of the kelvin, with 0 °C corresponding to 273.15 kelvins, approximately the melting point of water under ordinary conditions.

The problem with this description is that it does not make any distinction between the kelvin unit of temperature versus the Kelvin thermodynamic temperature scale, which measures temperature relative to absolute zero. For example, take the statement that "zero kelvins is absolute zero". Consider the question: what temperature change is necessary in water at standard temperature and pressure in order to liquefy it? The answer is "zero kelvins", but this most certainly does not indicate that the liquefied water would be at absolute zero.

My version of the definition used terms and phrases introduced by the BIPM brochure:

1. a kelvin is 1/273.16 of the thermodynamic temperature of the triple point of water,
2. it is common to measure relative to 273.15 kelvins, and
3. such practice is referred to as "Celsius temperature".

It also includes some embellishments I thought were fairly benign (and were pre-existing):

• absolute zero is where (not "when") molecular motion stops, and
• 273.15 kelvins is approximately the melting point of water.

All in all, I think this is a fairly uncontrovercial definition of the kelvin, and I'd like to reinstate it, or something like it. --P3d0 03:30, Dec 30, 2004 (UTC)