Talk:Thermal conductivity

In the section "First definition," the article states that "To physicists [1], thermal conductance is ..." and the footnote [1] points to the well-known and authoritative NIST Special Publication 811 which actually does not define or use the term "thermal conductance" anywhere. This link is not really good evidence or corroboration of the assertion that there is a standard definition of thermal conductance. (It is, however, a good place to find authoritative definitions of thermal conductivity, -insulance, -resistance, and -resistivity.) Perhaps it should be moved?

Other semi-authoritative references like the CRC Handbook of Chemistry and Physics, 2002-203 edition don't define thermal conductance either. (Although they do have conductivity.)

Does anyone have a more authoritative definition of thermal conductance as having the dimensions W / (m^2 K) (which can be more generally stated in terms of SI base units as: kg / (K s^3) ) Preferably from an authoritative standards body, and not some individual's webpage or questionable textbook.

I have a feeling that "thermal conductance" is a not-well-standardized term, if it's proper at all.

I also find the phrase "to physicists" very questionable, as if physicists don't or shouldn't follow the SI but rather have their own standards. (Last time I saw this in Wikipedia, it was a rather confused physicist arguing that "in physics," a pound was a measure of force, not mass.) This phrase should probably be removed.

By the way, I'm interested in the right terminology because I'm building authoritative references in my programming language/calculating tool Frink (http://futureboy.homeip.net/frinkdocs/).

No, the NIST doesn't define thermal conductance, but as I recall it does define electrical resistance and conductance as reciprocals of each other. I felt I was on pretty safe ground assuming thermal conductance was the reciprocal of thermal resistance. Thats also how it works in "building trade" units.

I've since added a few more web references that look fairly reliable: [1] (http://gscassociates.com/wg8/edcs/text/unit.html) and for the future[2] (http://www.iso.ch/iso/en/ittf/PubliclyAvailableStandards/ISO_IEC_18025_Ed1.html). The first says that the terms I've used - including thermal conductance - are in ISO 31-4:1992, Quantities and units - Part 4: Heat. Another new reference ([3] (http://www.npl.co.uk/thermal/faq_index.html#heat%20transfer%20property), thermophysics FAQ 5) is much better for "builders" units.

I've replaced "To physicists" with "For general scientific use". I would have put "in SI units" or something similar, but I think maybe only some of these units are defined as part of the SI, while others are derived. After all, "bulider's" thermal resistance is loosely "in SI" if it's expressed as (K m^2)/W. And for SI we should really refer to the BIPM rather than the NIST.

Andy G 18:25, 10 Sep 2003 (UTC)

Thermal Conductivity definition ok?

I was wondering if someone could explain the definition of thermal conductivity at the top of the page:

(thermal conductivy = heat flow rate / (distance x temperature gradient).

It does not appear to agree w/ several other places I have seen the term defined as:

thermal conductivity = (heat flow rate x distance)/ (Area x temperature gradient)

I am not an expert in the field, and may be missing something - can anyone explain the discrepancy?

You are right, and this equation is wrong. This definition is also wordy and vague. I propose the following edit:

In physics, thermal conductivity, λ, is the Intrinsic property of a material which relates its ability to conduct heat. It is the quantity of heat, H, transmitted through a unit thickness, d, in a direction normal to a surface of unit area, A, due to a unit temperature gradient, ΔT, under steady state conditions and when the heat transfer is dependent only on the temperature gradient.

thermal conductivity = heat flow rate × distance / (area × temperature gradient)

λ = H × d / A × ΔT

I also propose changing thickness, T, to thickness, d, in the equations later in the article. This choice of variables avoids confusion with temperature, T, and temperature gradient, λT.

If no one objects, please update the wiki. Comment posted 5/26/05

Example values for common elements OK?

I notice that the Thermal conductivity for Silver and Copper is given as 406 and 385 W/(m*K) on this page, but is stated as 429 and 401 W/(m*K) on those elements' own pages. What is the ultimate authority on these numbers? --Drhex 11:21, 2005 Jun 10 (UTC)