Talk:R-value (insulation)

Incoherence in SI units
The tables in the "Example values" section show a column supposed to be in SI units, but the units contain inches (m^2·K/(W·in)) instead of metres (conversion factor: 0.0254 m/in). I have seen the warning, but I think it is pretty useless to present mixed units since people who want SI units won't use inches anyway. Why not convert the figures by dividing them by 0.0254? —Preceding unsigned comment added by Nounec (talk • contribs) 11:07, 26 October 2010 (UTC)


 * I absolutely agree. Who the hell thought that it would be a useful idea to give U-values in SI units per inch? —Preceding unsigned comment added by 137.222.122.47 (talk) 15:18, 1 December 2010 (UTC)


 * I just tripped over this one. I am glad it is the first item on this discussions page! I volunteer to make the table consistent. Henk.muller (talk) 19:40, 22 December 2010 (UTC)


 * Help, somebodyh changed the text on the units, but not the values. Something is going horribly wrong? See Revision as of 23:16, 2 January 2011, by Alexibu. 77.99.40.19 (talk) 21:56, 12 January 2011 (UTC)


 * I couldn't agre more with this. I don't think I've ever seen any scientific measurement with dimensions m*m*K/(W * inches)! I did a double-take and actually searched the page to see if "in" was defined in some other weird way because I just really wasn't expecting anything as odd as this. It's confusing as hell and mixing imperial and metric units is the sort of things which causes spacecraft to crash. What hope has the average man trying to figure out how much insulation his house needs? — Preceding unsigned comment added by 86.137.76.104 (talk) 12:16, 15 June 2012 (UTC)

This is still here!!?? If I can, I'll delete it. Bradrh (talk) 22:33, 11 February 2015 (UTC)


 * The table and it's "mixing" of SI and non-SI units are correct. In the article's own words, the table provides "R-values per inch given in SI and Imperial units".


 * The table gives thermal resistance values in both imperial and SI units for various materials. The thermal resistance of a material depends on thickness. Presumably for ease of comparison, the resistance given for each material is for a one inch thick slice. The material is one inch thick irrespective of the system of units used to convey its resistance. If the first column were converted entirely to SI units then it would be concerned with resistivity rather than resistance.


 * I'm guessing that 'R-values per inch' was used in preference to 'R-values for one inch' to avoid giving the impression that the resistance was measured on an actual one-inch thick sample of material.


 * I will undo the deletion of the table. Pololei (talk) 07:46, 28 February 2015 (UTC)

Please do not, Only folk in the USA and UK residents over 50 know what an inch is. NO ONE wants metric per inch A more useful thing might be U values for 100mm thickness if you must. — Preceding unsigned comment added by 92.6.164.252 (talk) 00:27, 13 January 2020 (UTC)


 * The RSI-value column was likely intended to present values per inch because its contents, I'm guessing, were derived from those in the R-value column. (The R-value presumably orignated as a concept in the U.S.—which explains the choice of the inch in the table—with the RSI-value being a derivative.) Presentation of RSI-values per 100 mm may be more appropriate. However, RSI-values per a metric unit length are already given in the last column in the table.


 * To make it less conspicuous, and to help avoid objection in future, I've removed the inch from both the R-value and RSI-value column headers: strictly speaking, the table simply presents RSI-values and R-values, albeit normalised to a certain thickness, rather than RSI-values and R-values per unit thickness. Pololei (talk) 02:56, 14 January 2020 (UTC)


 * I think the fact that the column headings don't include proper units make it more confusing. At present, the RSI columns both claim to be in units of (m.K/W) but only one of them really is. Can someone please change the units in the RSI per inch column to the true units, ie (m^2.K/W.in) I would do it myself but don't seem to have the option. —Dom.uk.1 (talk • contribs) 08:21, 27 April 2020 (UTC)

why does the article interchange between watts and BTUs?
One is a unit of power, the other is a unit of energy. — Preceding unsigned comment added by 84.13.65.56 (talk) 06:23, 15 September 2020

Another impossibility
"In the absence of radiation or convection, the surface temperature of the insulator should equal the air temperature on each side."

This is impossible. If you visualize thermal gradient as a voltage drop, there is obviously a thermal difference when using a conduction model of heat transfer. —Preceding unsigned comment added by 146.145.138.42 (talk) 20:21, 25 August 2010 (UTC)


 * Well there is never an absence of radiation or convection, but in the usual case it is a thin layer of non-moving air near the surface that limits heat transfer. In the absence of that layer, the boundary condition is as indicated. Gah4 (talk) 10:56, 22 January 2019 (UTC)

How does the rate work without factoring time into the equation?
How does the rate work without factoring time into the equation? In the formula example used the answer is 5 watts per sq m. That doesn't mean anything without time.George Swantner 18:55, 14 September 2007 (UTC)
 * One watt is equivalent to one joule per second. The measurement and subsequent calculation (see ASTM C177 for more info) is based on heat flux (q - heat loss/gain per unit area per unit time), is taken while the sample is at steady state, meaning that there is no change with time, or $$\frac{\partial q}{\partial t}=0$$.  I hope this helps.  Bill Fabrizio --65.206.71.3 (talk) 17:48, 21 January 2009 (UTC)

I've taken the liberty of editing this, even though it's a topic on which I'm no expert. However, it's very clear that at least in New Zealand, R-value is taken to mean SI R, and nothing to do with fahrenheit or BTU. Perhaps the most contentious bit is my argument about thermal conductance through ceilings, so if any physics or building experts are out there, and I haven it terribly wrong, accept my humble apologies.--Limegreen 04:22, 26 May 2005 (UTC)


 * To avoid confusion, I think Wikipedia should default to SI units. 217.121.144.89 09:23, 28 May 2005 (UTC)


 * Good changes. I didn't want to appear like some avenging crusader, which probably impaired my clarity.Limegreen 00:17, 29 May 2005 (UTC)

F° versus °F
I think it is more proper to use the unit F° rather than °F when referring to temperature differences. One does not see this distinction made very often but it is technically correct. °F refers to a position on a temperature scale while F° refers to a unit of measure. Of course the same holds true for °C and C°. —Preceding unsigned comment added by 12.10.81.212 (talk • contribs) 2005-11-09 16:19:05


 * I don't think I've ever seen that notation used, but it sounds like something that could be very useful. Can you show any examples of where that notation is used or defined? It's rather difficult to search for on Google, since the degree symbol just gets stripped. &mdash; ?ospace; font-variant: small-caps;">HorsePunchKid &rarr; &#x9F9C; 21:27, 9 November 2005 (UTC)


 * I've never seen that notation used, neither for the Celsius nor Farenheit scale. A temperature difference is in degrees Celsius; it doesn't make sense to talk about Celsius degrees. If this (distinction) is technically correct as claimed, please state your source. At least for my own part, as an energy engineer, I've never seen it in use. As an alternative to °F or °C, all temperature differences could be stated in Kelvin [K].


 * I have seen that distinction used, most notably in one or two print encyclopedias (World Book? Encyclopædia Britannica?).  But it is not "technically correct".  It is the type of thing that the CGPM partially endorsed back in 1948 (but their since-abandoned "solution", never widely followed in any case, was to use "deg C" to mean something different from °C), but they definitely disavowed that a long time ago, I think in the 1967-68 CGPM which also changed the name of the old "degrees Kelvin" to "kelvins", making clear that they are the same units, same symbols, whether measuring a temperature reading or a temperature difference.  Gene Nygaard (talk) 05:14, 9 December 2008 (UTC)

I was taught this distinction in my high school chemistry class. (And no, I am not the original poster on this topic.) Jasper (talk) 15:52, 26 July 2009 (UTC)

For consistency with the rest of Wikipedia and because all the cited references for the article use the notation °F, I have changed the few remaining instances of F° to °F Elfstrom (talk) 16:14, 9 August 2009 (UTC)

There's a good reason most of us have never heard of this "F-degrees". The F scale and the C scale already do implicitly imply, distances on a number line. That is why, there is a temperature called "zero degrees": so that the axis has an origin. alanmark@breakthru.com —Preceding unsigned comment added by 121.97.59.62 (talk) 08:08, 21 March 2010 (UTC)

Incredible external link?
Granted, it may be difficult to get a nonprofit link which is informative in such a mundane (and yet profitable) subject, but I do think that the "'R' fairy-tale" site, as hosted by an organization which seeks to profiteer off of an architectural concept (the "Monolithic Dome"), is not suitable to be linked to by an encyclopedia. Aside from discussing the "ignored" factor of convection on heat flow, the Monolithic Dome Institute does nothing other than to promote the sale of all things "Domephile." And as their financial future is dependant on their ability to promote dome-mania while dismissing weaknesses in their design, the institute does not qualify as a nonbiased source. I would recommend the replacement of those two external links with either non-profit organizations or academic sites. — Preceding unsigned comment added by 68.117.6.103 (talk) 07:26, 30 March 2006

"Some other countries" = "U.S."?
I think it would be useful to make it clear that the "some other countries" that use the BTU definition of R-factor include the United States. Also, I think it would be useful to give a straightforward defintion of the U.S. version of R-factor in W/m2KPciszek 19:15, 15 March 2007 (UTC)

Perhaps more accurate to say "the only country that persists with the BTU definition of R-factor is the United States". — Preceding unsigned comment added by 61.69.223.50 (talk) 12:11, 11 July 2019 (UTC)

Company Pink
I think of great value would be a section on how this law about the R-Value being the standard for heat transfer for insulation in the US came about. About how "Company Pink" made a deal with congress to get this law passed (which had an insulation with very good R value but quite low conductive and radiative values) which made R-Value the standard. Also how the R-Value rule is keeping many good competitors from being profitable because they have lower R values but much better values for the other two types of heat transfer. I can't give citations on this, but my physics professor ranted for ten or so minutes on it. I'm sure some of you have heard of this. Ergzay 03:12, 21 September 2007 (UTC)

Reference spam
It doesn't seem that the single reference link (PDF file) contains any verifiable information and is not directly related to any information on the page (in fact, it contradicts much of it). Moreover, the target of the link is on a commercial site and clearly promotes a product.

—Preceding unsigned comment added by 68.147.142.222 (talk) 19:29, 12 October 2007 (UTC) The photo and reference 1 to aerogel work is completely superfluous and is essentially an advertisement for the LBL researcher's work. Both the figure and reference should be deleted, or at a minimum moved to a section discussing research on advanced materials for high R-value insulation. It is not of primary importance to understanding what an R-value is. — Preceding unsigned comment added by 71.172.236.87 (talk) 16:46, 11 June 2022 (UTC)

Discussion of what it includes
I just removed this discussion from the article, thinking that it belonged on the talk page. So here it is:

"Correction! The transfer of heat by conduction is specific ONLY to a material's specific resistance to that transfer. The only fixed factor is the material's density which in turn determines its resistance or "R" value. ONLY If the material is porous, could the the R value include convection or radiant transfer potential. The National Fire Service describes CONDUCTION as "heat transfered between two solid surfaces in DIRECT contact with each other that allows heat to be transfered between physical objects." By definition, this would exclude radiant and convective transfers as these means are based on radiance or particles in cirrculation. Simply because the " thermal Insulation Community" dislike competitive products that make insulation more effective, is no reason to rewrite basic scientific principles. "

I think the above is unclear and confused and inappropriate style for wikipedia, but thought it might be more productive to put it here than just delete it. It's true that most solid materials don't have convective or radiative heat transfer inside them, but most insulation materials are not homogeneous solids. I'm also not sure the relevance of the NFS definition of conduction.Ccrrccrr 04:26, 16 October 2007 (UTC)

Apparent thermal conductivity
It would be useful to have the analytical method or model used to determine the 'apparent thermal conductivity' of the material quoted by Yarbrough. Is this 'standard' of quoting R-value universally accepted or is it a proposed industry standard. Is the 'r-value' obtained useful for the range of applications and climates in which it is used,


 * An opinion related to this that was in the article, but belongs hereCcrrccrr (talk) 13:50, 14 December 2007 (UTC)


 * Ideally, a rating system should properly describe and separately quantify the heat transfer depending on its mode of transfer e.g. conduction, convection or radiation (see below) as they are processes that can be independent of one another.

Impossiblility
I just deleted this false statement from the article:


 * It is impossible for one type of material to effectively retard all types of heat transfer.

This is not true. A block of foam blocks airflow effectively, retarding convection dramatically. It has low thermal conductivity, thus reducing conductive heat flow greatly. And it's opaque to thermal radiation. Almost any material is opaque to thermal radiation--windows for thermal radiation need to be made of special exotic materials like ZnS. See, for example,. "Radiant barrier" materials prevent radiant heat transfer through the airspace adjacent to a wall, rather than preventing radiation from going through the wall. Thermal radiation doesn't go through the wall, unless you make a wall out of translucent polyethylene.Ccrrccrr (talk) 14:01, 14 December 2007 (UTC)

Many materials be opaque in the visible light spectra but not in the infra-red and very few to x-rays. Materials can be spectrally selective and emissivity will vary according to wavelength. Even the sun does not emit equally at all wavelengths - that's why it appears yellow. If an object can absorb thermal radiation then it will also emit it. Therefore, they can 'appear' to transmit it. Radiant barriers work partly because they reflect radiation to its source but mainly because of it low emissivity, they won't re-radiate it on its opposite surface. A clay brick is different - it progressively heats up and will radiate its heat into the interior. 219.90.145.245 (talk) 00:12, 15 December 2007 (UTC)


 * X-rays are much shorter wavelengths than visible; IR is longer wavelengths than visible. The analogy to X-rays is irrelevant.  Yes, some materials can be opaque to visible and transparent to IR.  But they are hard to find, especially for the thermal (far) IR band.  Check out the link above.


 * Emissivity does vary widely. Yes, high absorption means high emissivity as well.  Yes, an object can "appear" to transmit radiation because it absorbs it and re-radiates it--a simple wood wall or a pane of glass are both example in which this would happen.  The heat transfer is reduced, but not a lot.  That's part of the classic greenhouse effect.  Yes, a radiant barrier can work either by reflecting radiation to the source or stopping the re-radiation.  However, if a solid object is absorbing radiation on one side, and re-radiating it on the other side, the heat has to get from one side to the other.  It does so by conduction.  If the solid object includes bulk insulation between the two surfaces, that re-radiation and apparent IR transmission is reduced by the r-value of the bulk insulation.  The bulk insulation is not somehow bypassed by the radiation.Ccrrccrr (talk) 01:22, 15 December 2007 (UTC)

No argument here but the surface temperature of your insulator (which is exposed to solar radiation) is going to be much higher than the ambient temperature). R-values being constant and temperature gradient same, the surface temperature of the other side of your insulator will also be high (and also not equal to the interior air temperature).  This hot surface will be the source of thermal radiation.

The easiest way to test this is create a foam box. Put a infra-red camera inside it and stick the box in the midday sun. See what kind of thermal radiation is being transmitted. —Preceding unsigned comment added by 219.90.145.245 (talk) 03:46, 15 December 2007 (UTC)

Overwhelming bulk insulators
Another section removed:
 * Similarly, bulk insulators can be easily overwhelmed by solar radiation during summer (where black globe temperature far exceed ambient air temperatures but are much more effective in reducing winter convective losses. Often a judicious combination of both materials is required to achieve an optimum solution for a mixed climate.


 * WBGT is designed for assessing heat stress for people who are outdoors. Figuring out how to apply it to buildings is not straightforward and would be original research.  That sounds useful, but it doesn't belong in Wikipedia.Ccrrccrr (talk) 14:15, 14 December 2007 (UTC)


 * Wet bulb globe temp is not the same as black globe temp. My point being that when insulation is modelled, is it being subjected to the same radiation flux as would experienced during summer.  Furthermore, is it measuring the temperature of the ambient air on each side of the insulator or the temperature that is also due to the effects of radiation (black globe temp or mean radiant temperature).  This would markedly change the results of any formula on 'thermal conductivity' that you use.Dymonite (talk) 10:00, 23 December 2007 (UTC)

Right--WBGT and BGT are not the same, which was a problem with that original text which linked to WBGT using the name BGT. And yes, a black building in the sun will have an average wall temperature similar to BGT, which will be higher than the ambient air temperature. The implications of this are that light colored exerior finishes, particularly those certified to reflect solar IR (see cool roofs) are advantageous and that more insulation might be needed than what would be calculated based on ambient temperature. That additional insulation may be of any type--radiant barrier or bulk--the high BGT does not have any implication for which type is preferred. Perhaps I should have corrected the paragraph to explain all that rather than deleting it.

Foam insulators
It seems that ccrrccrr is a strong advocate for foam insulation and it seems the thread is arguing for its use. I think it sounds like a promising alternative. But this product is only new to my country and with most insulation products, you cannot get any info on the lab methodology. What it the emissivity of foam for heat energy? Has this been determined experimentally. If I stood under a block of foam in the midday sun, how much of a temp difference can I experience?

219.90.145.245 (talk) 00:12, 15 December 2007 (UTC)


 * If you want to know my opinion (something I try to keep out of my editing), I think foam is great for some things but I'm much more enthusiastic about cellulose, because you can get a much higher R-value for the same expenditure with cellulose, and I think that the global warming potential of HFCs in many foams are a serious problem that should be taken more seriously particlarly by people who advocate superinsulation to reduce CO2 emissions.


 * I certainly hope my edits don't turn this into an advocacy piece for foam--and I hope that they don't undercut the real value of cool roofs and radiant barriers. My goal is accuracy, not advocacy for any particular product.  I don't object to claiming benefits for cool roofs and radiant barriers--I just want to make the basis for that claim technically sound.  And I want to keep this page a discussion about R-values, with mention of and links to other topics as appropriate.


 * I'll try to answer your questions soon.

Ccrrccrr (talk) 13:07, 15 December 2007 (UTC)

If you look at the building insulation materials page and the history of my edits there, I recently restored some disadvantages of foam that had been listed and were removed by someone else. No, I'm shilling for the foam industry!

I don't know the emissivity of foam, but I think it's a safe bet that it's high, like around 0.9. If you used a piece of foam as a porch roof, the top surface of it would get hot from the solar radiation. But you'd want to cover it with something anyway, since it's not resistant to UV damage. If you do that, then it's the emissivity of that cover (paint, shingles, metal, etc.) that matters on the top, and the emissivity of the top surface of the foam doesn't matter. If the top surface does get hot from radiation (black shingles, for example), most of the heat it absorbs will be lost by convection at the top surface, and relatively little will conduct through. The temperature on the porch will not be affected significantly by the solar radiation on the roof--most of the heat will come in the open sides of the porch.

If we wall in the sides of the porch, and insulate them, then the heat gain from the heat flow through the insulation on the roof will become important, especially if, with black shingles, the top surface is hot. Then we might want to reduce that heat flow. We could do that with any bulk insulation (more foam, or cellulose which is much cheaper), or we could do it with a radiant barrier on the bottom of the foam. In a hot climate like most of Australia, a radiant barrier would be a really good choice, because it will give significant benefit at low cost. If you are in a mixed climate, it's less clear what you should do--if you add more foam or cellulose, that will help in both summer and winter, whereas if you add a radiant barrier it will help significantly in summer, but not much in winter. That's because the radiant barrier helps for downward heat flow much more than for upward heat flow.

I hope that helps answer your question.Ccrrccrr (talk) 14:28, 15 December 2007 (UTC)

Opinions about what data sheets should include
I removed the following from the article because it is an opinion, which doesn't belong in Wikipedia. I agree with some of it, but that's irrelevant. Opinions belong on the talk page.Ccrrccrr (talk) 13:09, 15 December 2007 (UTC)


 * Data sheets for insulators can be scrutinised for appropriate testing methodology. Such a analysis should specify the type and intensity of heat energy that come from both conductive and radiative sources.  It should also document both ambient air temperature and surface temperature on both sides of the material.  It should also make clear other specifics pertaining to details about convective gas/fluid movements.  A calculated value for thermal conductance and emissivity can then be derived from the data set to quantify the impact of both conduction and radiation through the material.

Type of heat source used
I removed this sentence:
 * The measured heat transfer will vary depending on the type and intensity of heat source used in the model.

If you have a conventional bulk insulator, the heat transfer through it comprises radiation, convection, and conduction. The mix of those three does not depend on the mode by which the heat arrives at the surface of the test box. Once the heat arrives at a wall, it is just heat, regardless of how it arrived there, and it's free to proceed through the wall by whatever means is permitted by the insulation there. It's not like heat that arrives at the wall as radiation can only go through the wall as radiation.Ccrrccrr (talk) 13:15, 15 December 2007 (UTC)

Definition of R value
I added:

R-value is a measure of apparent thermal conductivity, and thus describes the rate that heat energy is conducted through a material or assembly, regardless of its original source.

The higher the R-value, the slower at which heat energy is conducted. It describes the rate of heat transfer and not the total amount that is transferred. It also does not consider the proportion of radiant energy that is reflected away from the surface of a material before being transmitted through it.


 * Heat energy does not travel through a bulk insulator by radiation. Mostly by conduction and a little by convection (if there is a gas/fluid medium within).  Not all radiant energy will be converted into conducted heat - some of it reflected away.  Materials with low thermal conductivity only retard the flow of energy (most of it eventually reaches its destination).  The most effective means of reducing solar radiation is to reflect a portion of that energy away rather than having a massively thick bulk insulator to deal with the load to slow its passage.

Could you please sign you comments on the talk page? It's hard to follow the conversation otherwise. I can't tell whether the above is two people talking or one. ThanksCcrrccrr (talk) 23:31, 30 December 2007 (UTC)


 * oops sorry

Dymonite (talk) 02:26, 31 December 2007 (UTC)

We might want to add material on the mechanisms of heat transfer through insulators on the building insulation page--it's a surprisingly interesting topic. Heat does transfer through an conventional bulk insulator by all three mechanisms. A bulk insulator is a large collection of little air pockets. For the moment, imagine a closed-cell foam. Heat can:
 * Flow through the solid material, around the air cells.
 * Flow across one air cell, then through the solid cell wall to another air cell, across that air cell, etc. The heat transfer across the air cell can happen via:
 * Conduction through the air (air's a very poor conductor, but it does conduct some).
 * Convection loops within the air cells (hard if the cells are small, but can happen)
 * Radiation between the walls of the cell.

If the foam is not closed cell, or the insulation is fibrous or loose fill, there can also be larger-scale convection permeating the insulation.

In the ASHRAE handbook there's a great plot of apparent thermal conductivity as a function of insulation density--at low density the thermal conductivity goes up due to more convection, whereas at high density the thermal conductivity goes up due to conduction. There's a point in the middle where thermal conductivity is minimized.

I don't know the exact relative contributions of the three mechanisms in different insulation types--it's hard to make measurements that separate them. (You can do stuff like measure the insulation in a vacuum--if it's not closed-cell--and eliminate convection, so you are just measuring radiation and convection. Then you can figure out how much is rad vs. conv. by looking at the dependence on temperature.)  But the nice thing about apparent thermal conductivity is that you don't need to know how much of the transfer happens by what mechanism. If you know the R-value and "you know the temperatures on the two sides," you can calculate how much heat flows, while totally ignoring the mechanism of how it gets through the insulation.

The issue you are talking about regarding reflecting solar radiation away is addressed in the cool roofs article. That is part of what determines the temperature of the surfaces on the two sides of the insulation. That's really important (italicized above). Maybe we need to emphasize that more in the article.

Then there's the issue of rate vs. total amount. To get from rate to total amount, you integrate over time. Ccrrccrr (talk) 13:35, 31 December 2007 (UTC)


 * I agree that an article on cool roofs and radiant barriers are an important counterpoint to a discussion of R-values.  The literature is peppered with R-values and nothing else.  The overemphasis on retarding heat transmission through materials diminishes the equal importance of reducing a radiant heat load in the first place.  Whilst R values have no place in the description of radiant barriers, there needs to be better standards to describe their usefulness in situations of high solar gain. Better consideration of terms like emissivity and reflectivity need to be considered in the building literature.  I think a clear distinction should be made in the article between 'ambient air temperature' and surface temperature.  These are two different things.  Bulk insulators behave differently with an air temp gradient of 0C (outside) and 20C (inside) vs 20 (inside) and 40 (outside).  The air temp gradient may only be 20C in both cases but the surface temp may quite different during summer due to the effect of thermal radiation within the roof space.

Dymonite (talk) 01:33, 1 January 2008 (UTC)

Dymonite (talk) 01:33, 1 January 2008 (UTC)

This entire page is extremely confusing and contradictory, and I have a PhD in materials engineering. In the first section, you define R-factor as the temperature rise per unit of heat flux (which is bizarre anyway, but which I'm fine with accepting if the consensus is that this is the definition). Then down in the examples, you give R-factors in units that are in my opinion actually correct (if weirdly mixed between SI and non-SI); namely, the fundamental material property that relates cross sectional area, thickness, temperature difference, and total heat transfer (not flux). I looked dozens of places online to try to find anyone else who defines R-factor this way, and came up with nothing. Everyone else uses either K/(W-m) or Hr-F-ft^2/BTU-in, making it equivalent to thermal resistivity.

Defining R-factor in the way it is done here is very odd; it means that the R-factor is different depending on cross-sectional area. That's just outright strange and non-intuitive, and I've never seen it used in that fashion. Usually the R-factor is a fundamental property of the insulation independent of area and thickness.

Anyway, I don't have the time to fix this, and don't want to in case I'm wrong (which I'm fairly sure I'm not, but maybe people in construction do things very strangely), but I wanted to at least point out the internal inconsistency and overall very non-standard definitions. — Preceding unsigned comment added by 50.78.2.197 (talk) 13:38, 24 September 2012 (UTC)

"Reader's Comment"
''Reader's comment: I Subscribe to the U.S. FTC prohibition to the use of R-value per inch. So this section "Values per inch" is not appropriate (in the USA). Tables or graphs should be developed based on specific available products.''

(not my comment--I'm just the editor the moved it here from the article. Note that the article has a discussion of this issue in the section above the list of R-values.Ccrrccrr (talk) 03:22, 8 January 2008 (UTC))

I'm not sure if this is the appropriate place to comment, however, would it be possible to include units in the tables? Yes, it is possible to find them from reading the sections above, but for clarity and completeness it would be useful to have the units explicitly stated in the table. Kmcfar2409 (talk) 15:43, 24 February 2008 (UTC)

Refernce errors
This recent edit changed some of the references and left both conflicting (autogenerated1) and undefined (autogenerated3) references. I'm just passing through and don't have the expertise to straighten these out. -- Tcncv (talk) 04:01, 31 October 2008 (UTC)


 * Thanks--I think I fixed it but I encourage people to check. —Preceding unsigned comment added by Ccrrccrr (talk • contribs) 23:29, 31 October 2008 (UTC)

R-value includes radiation
The definition wrongly states that R-value doesn't include the effect of radiation. Insulating batts with foil coatings achieve a higher R-value because of the role the foil plays in reflecting long-wave radiant heat eg from hot roof tiles, so R-values most definitely do include radiation effects. R-values are about heat transfer by whatever means from one control volume to another. Jdpipe (talk) 01:59, 16 November 2008 (UTC)

R value is defined by the engineering equation where it is used to calculate conductive heat transfer. The equation that uses R value definitely does not calculate thermal radiation heat transfer. While commercial insulation products may have features intended to reduce radiation heat transfer (reflecting thermal radiation, emissivity), this effect is not addressed by the heat transfer calculation that utilizes R value. R value is a measure of conductivity only. Additional factors can be included in the same heat transfer calculation to address convection at the boundaries, but calculating radiant heat transfer requires an entirely separate calculation that does not use R value. KPM PE 11/21/2008 —Preceding unsigned comment added by 159.214.124.1 (talk) 16:08, 21 November 2008 (UTC)


 * Dubious. The R-values cited for window glass are dominated not by conductivity but by the convective barrier due to natural convection at the air surface. Therefore R-value is not defined only in terms of conduction. QED. Furthermore, windows with 'low emissivity' coatings are given higher R-values. Hence radiation is also encapsulated by R-value. Jdpipe (talk) 12:54, 24 November 2008 (UTC)


 * This discussion is a bit like discussing what kind of bass you have on your angle when you don't even know if it's a fish at all. The source of the confusion is that R-value is very ill defined to begin with. There isn't even agreement on whether it is a physical quantity at all - or rather a unit. See Talk:Thermal conductivity. For me, coming from a physics background, this is very unsatisfactory, but I guess it works for the construction industry somehow. &mdash; Sebastian 17:35, 25 November 2008 (UTC)

Practical significance
(Headline inserted by me. The following discussion was a continuation of the previous one. It primarily serves to improve our understanding of the topic and is not intended to effect any immediate changes to the article.)

The bass analogy is a good one however R-value does not actually "work" for the construction industry. Since R-value is by definition "conduction only" using it to determine proper construction for energy efficiency makes about as much sense as buying and selling land based on one dimension. No one will purchase a property that is 100 x whatever, and yet we pass laws and make rules involving R-value as if it were somehow actually representative of the thermo that is taking place. R-value is important, as is the "100" in my example, yet proper decisions about the energy transfer or the property cannot be made without all of the information. What is being called the Effective R-value does include the net of conduction, convention and radiation as it is empirical. R-value is theoretical. (Cam09RFWWW (talk) 13:20, 7 January 2009 (UTC))


 * Are you arguing for your view that "R-value is theoretical" to be included in the article, or are you trying to understand the concept? If the former, then the answer is simple: Provide a reliable source. Even if everybody here on this talk page agreed, that wouldn't be a reason to change the article - it would only be original research. As long as we have lots of sources, from industry fact sheets to recommendations by the US Department of Energy, it's obviously practical. &mdash; Sebastian 01:23, 9 January 2009 (UTC)

I apologize if my comments are not exact for the section. My entry above was my first visit to Wikipedia. Anyway, your comment about the building industry and R-value really caught my attention. I actually believe that R-value is quite a problem in and of itself since many energy related decisions (and laws) are made with it as a basis and the fact that "R-value is very ill defined to begin with" leads many architects and engineers to make poor decisions. If people believed that radiation is somehow encapsulated by R-value and/or convection is somehow encapsulated by R-value, we would have exactly what we have today - a real problem. I don't know how to fix this but I can build a house with a very low R-value that outperforms the same house with a very high R-value. Thus, it is my opinion the use of R-value as it stands today in the construction industry is is almost bogus. The industry needs a new value that is representative of the full heat transfer taking place (like an assembly value) that is more appropriate for real world construction. I will look for sources. (Cam09RFWWW (talk) 16:52, 10 January 2009 (UTC))


 * That's good - no need to apologize. I have to apologize for not realizing that you were new.


 * I can learn a lot from this conversation, too. I'm not an expert in American construction, so you probably know more about the practical side. Would you have any example for poor decisions by architects and engineers caused by this? It's hard for me to think of any: total R-value is (only) conductive R-value for walls, so a confusion there doesn't matter. For windows and doors, the situation is complicated because it depends on usage. If users always keep blinds or shutters closed, then they will experience less heat loss by radiation. And if they open them often, they will get a lot of convective heat transfer. But you can't blame the architect for that, now can you? &mdash; Sebastian 00:44, 11 January 2009 (UTC)

Okay. I will need a little time to prepare this response properly and come back to it. For now, at the core, laboratory tested materials for walls experience only conduction as one would expect since R-value is only about conduction by definition. Therein lies the problem, real-world walls have conductive currents that dramatically influence the perfomance of the wall and radiation also has an effect on the performance of the real world wall. Thus a wall material that performs to R-19 in the laboratory might perform like an R-8 wall in the real world. For one thing, architects talk in terms of the insulation material (maybe fiberglass bat) as if this is the R-value of the wall and do not count the wood that connects 20% of the building interior to exterior (wood has a very poor R-value). This is being called R-performance by some, making things that much more confusing since design is based on conduction but reality suggests that we attempt to tie the rest of the thermodynamics to it in some way. Things get more complicated when you consider a wall system like Insulated Concrete Form that is about R-22 in the laboratory but performs as high as R-50 in the real world. As you know concrete has virturally no R-value at all. As far as I know, thermal mass is usually completely ignored. Architects and Engineers design to R-value and they should not. (Cam09RFWWW (talk) 23:34, 11 January 2009 (UTC))


 * I see what you have in mind now! The differences you're you're talking about have nothing to do with the difference between laboratory and real world, and practically nothing with convection and radiation. They are mainly caused by:
 * Good old industry cheating. It's the same as when you pay for 2 by 2 lumber, and what you get is actually 1.5 &times; 1.5 - 44% less! That doesn't mean that the industry uses different definitions for "size" or for "inch". (It may have to do with the fact that they are mostly men, but I don't want to spell out that silly joke here.) The ORNL measured e.g. for "R-19" fiberglass batts, even in the best case, only R-15.4 - 19% less than nominal. (The cheating may not be as bad; the difference may already include some thermal bridging effects, see below.)
 * Thermal bridging: This is expressed by a difference between "C-W R" (average of R-value of the cavity between the framing members) and "W-W R" (whole wall R-value, which considers the lower R/value of the framing material.) Differences are typically around 15-20%.
 * Improper installation. That can make a difference of 15% for fiberglass batts, and much more for loose fill.
 * Convection and radiation usually only become an issue with improper installation and when loose fill insulation settles.
 * (Values and definitions taken from Krigger/Dorsi, Residential Energy, p. 274). I haven't heard about the increase you mention for ICF - where did you read that? &mdash; Sebastian 20:44, 12 January 2009 (UTC)

ICFs perform significantly above their actual r-values per Construction Technologies Laboratory (CTLGroup) report entitled ("Analysis to determine the thermal mass performance of a typical 9-inch ICF wall" ). It discusses Effective R-value as well as geographical variance in R-value performance. CTLGroup, 5400 Old Orchard Rd, Skokie, Illinois 60077. I have the file and will eventually figure out how to load it or link to it. As I learn more about this I am convinced that, yes, it is "cheating" all right. If you use a hot plate lab method for r-value test (with static air) you can create the illusion that a lot of materials work well that really don't. Interesting that the paper on BAT insulation has no real impact on the laboratory test for conduction but in the real world the insulation will not work at all without it.(Cam09RFWWW (talk) 23:30, 16 January 2009 (UTC))


 * You're right, I forgot thermal mass in the list above. That typically leads to an increase of effective R-value of 30%-50% for heating (if you live in Minneapolis), and up to 130% for cooling (if you live in Miami) (Krigger/Dorsi, p. 274). But this is very shaky; it depends on pretty much every factor that you can think of, including user behavior such as when and how often do they open the windows or take showers, and many more. The situation is so complex that even the ORNL resigns and writes "There is no physical meaning for the term R-value equivalent for massive walls. ". If you really want to know how thermal mass affects your energy bill, you simply can not use any single value; you have to do a complicated simulation for a whole building over the course of a year. &mdash; Sebastian 19:44, 17 January 2009 (UTC)


 * I agree with Sebatian. It's very true that other factors affect efficiency other than insulation.  But it's silly to try to describe them all by "effective R-value".  That would be like saying that a low-aero-drag car has low "effective rolling resistance" instead of separately discussing aero drag and rolling resistance.Ccrrccrr (talk) 21:30, 17 January 2009 (UTC)


 * I may have not been clear if you thought I was a fan of "effective R-value" -- I see it as another way of trying to lump things into R-value (see beginning of all this where it was stated that R-values included radiation). We shouldn't say that it includes convection or thermal mass either. My point is that R-value does not help us build smart because it's only part of the picture. (Cam09RFWWW (talk) 21:15, 18 January 2009 (UTC))


 * Sorry imply you were a fan of "effective R-value". I was trying to state what I thought and contrast them with other options, and didn't mean to link them to you.


 * So now looking at where to go from here, I think that saying "R-value does not help us build smart" is an overstatement. It helps--it is a useful thing to consider, but it is just one of a large set of things that need to be considered.  Going back to my car analogy, both considering aero drag and considering rolling resistance are both helpful things to do in trying to make an efficient car, even though neither is sufficient (nor is the combination of just those two sufficient).  Just because one brick doesn't make a wall doesn't mean that bricks aren't useful for making walls.Ccrrccrr (talk) 23:21, 18 January 2009 (UTC)


 * I think we basically agree. My concern is that the industry doesn't seem to know that the other things are worth considering and that's a little dangerous.  Based on what I've been learning/reading etc. folks seem to want to mix them in together anyway!  You know, "the rolling drag factor" and whatnot.  I like your analogy but think buying property based on one dimension (eg. I'll see you a piece of land, it's a 100) is more like what's happening out there.  Like R-value, the 100 is important data and must be known, but no one in their right mind would buy a lot based on one dimension. Like the car, you can't make proper decisions regarding the car without considering both drag and rolling resistance.  Here, it's just a crazy to look at land without the 100 --- should someone want to sell a something x 200 lot. (Cam09RFWWW (talk) 00:03, 19 January 2009 (UTC))

Reported R-value for single pane window glass.
Modified by David Lambert, lambertdw@corning.com —Preceding unsigned comment added by 69.205.128.181 (talk) 21:03, 22 November 2008 (UTC)


 * The thermal resistance of the glass itself is negligible, as you probably know. The thermal resistance is all in the heat transfer from the room to one surface, and from the other surface to the outside....or vice versa for heat flow in the other direction.  That is how it is normally discussed.  It's analogous to the R-value of a whole wall assembly that includes those surface effects on inside and out, plus the wallboard, plus the insulation, etc., rather than being analogous to the value for a single material.Ccrrccrr (talk) 22:09, 22 November 2008 (UTC)


 * Consider three stage system comprised of convection, conduction, convection. The heat transfer rate across entire thing is keff*area*(Toutside-Tinside).  Because of steady state the convective transfers balance the conductive transfer balance the net rate.

These three equations are just right to determine the three unknowns: keff, and the surface temperatures of the glass.

Substituting gold foil properties has negligible consequence, as you say. Thank you for clarification. Agreement using this calculation is quite good. 199.197.135.180 (talk) 22:16, 2 December 2008 (UTC)


 * Which language/syntax are you using? &mdash; Sebastian 08:08, 28 December 2008 (UTC)


 * Mathematica, maybe? --DavidCary (talk) 13:05, 9 June 2014 (UTC)

Merge most of this page into "Thermal conductivity" and other articles
This article is more a source of confusion than of clarity. An analysis of this and related articles at Thermal conductivity shows that there is a huge overlap. I therefore propose to change this article basically to a short article as in the following box and move all other context into the appropriate articles.

I'm writing this from memory, but I've seen all of the above in reliable sources, except for #2, which I'm taking from this article, section U-value, and the discussion above. As for the remaining meanings, I feel it is cleanest if we:
 * 1) merge the description of physical quantities, such as thermal resistivity into the appropriate article, which currently is Thermal conductivity,
 * 2) merge the list into List of thermal conductivities.

Please keep in mind that this post is about the merge itself. I now realize that the list of different values might distract people into discussing that instead, but I would like to keep it to show that the term "R-value" alone can fill a whole article, so that there is an advantage in moving those parts out of the article that are better covered elsewhere. Please, let's first see if we can agree on the merge, before we delve into the details of the different definitions for "R-value". &mdash; Sebastian 03:05, 9 January 2009 (UTC)


 * Removed the merge request. R-value is a really important term, and needs clear discussion separately. There are aspects of R-value that are totally outside jsut thermal conduction, eg convection at boundaries. Jdpipe (talk) 01:52, 27 February 2009 (UTC)


 * I readded the request. The discussion is not suddenly over, just because you don't like it. Your argument does not address my concerns; as I pointed out, "R-value" is not just one term, but has a variety of quite different meanings, one of which is the same as thermal resistance. This ambiguity is not solved by just ignoring it. &mdash; Sebastian 02:50, 27 February 2009 (UTC)


 * Well, I waited a month before acting, right? Jdpipe (talk) 00:07, 3 March 2009 (UTC)
 * What? To the contrary! For a month, the request stood here without any objection. I could have closed it in my favor, but I thought it fairer to wait patiently. Then you come, write your objection here, and within minutes proceed to close the debate in your favor, without waiting for any reply. &mdash; Sebastian 16:11, 3 March 2009 (UTC)


 * Number 1 in the box is wrong. R-value is the thermal resistance of a unit area section, not total thermal resistance.  If you insulate a house with a surface area of 100 square meters with RSI-5 insulation, the thermal resistance is 0.05 degrees C per watt, not 5 degrees C per watt.Ccrrccrr (talk) 22:34, 28 February 2009 (UTC)
 * You are talking about the definition at Thermal conductivity, which is defined in analogy to electric resistance. What I referred to was Thermal conductivity. To be honest, I don't have a reliable source that cuts through this confusion. So, unfortunately, we don't just have to deal with confusion about the term "R-value", but also about the term "thermal resistance". &mdash; Sebastian 20:25, 1 March 2009 (UTC)


 * Every textbook I've seen defines thermal resistance as under Thermal conductivity. I think that using the same term (thermal resistance) two different ways is much more confusing than using R-value to refer to thermal insulance and only using the conventional definition of thermal resistance.  There's no source for the use of "thermal resistance" for thermal insulance under Thermal conductivity.  So that doesn't constitute a source we can use in support of this proposal.  Apparently you agree, so I'm not sure why you are proposing this.Ccrrccrr (talk) 20:56, 1 March 2009 (UTC)
 * I agree that there is a problem with the definition of "thermal resistance". But that should be discussed on that talk page. Regardless what definition for "thermal resistance" we agree on, the problem with "R-value" still persists. It is very easy to rephrase the box above so it fits your preferred definition: Just replace "thermal resistance" in item 1 with "thermal resistance of a unit area section". &mdash; Sebastian 22:11, 1 March 2009 (UTC)
 * That would help, but the section that it links to would need to be improved first.


 * Beyond that, I think that this article has a different focus and addresses a different audience than the thermal conductivity article. The only advantage I see of the merge is that we'd gain policing by some physicists to keep nonsense from getting on this page.  Ccrrccrr (talk) 22:35, 1 March 2009 (UTC)


 * Well, I'm already a physicist, and you seem to be good at it, too. I agree that the audience is different, and I'm actually not proposing a full merge; I would leave the text in the box. Do you feel there needs to be substantially more than that in this article? (I know, my proposal is a bit unusual since it is somewhere between a small article and a disambiguation page; I hope that doesn't cause opposition.) &mdash; Sebastian 00:26, 2 March 2009 (UTC)


 * R-value is not a physicist's term. It is a HVAC engineering term. The conventional usage is not the same as thermal conductivity. It is an overall thermal resistance including the effects of convection, conduction, and even to some extent radiation (although this is clearly problematic, with the T^4 relationship there). A suitable reference would be the ASHRAE handbook, I have the 1989 edition here, and that includes a definition which incorporates the effect of series convective and conductive resistances. R values are commonly used to label insulatin and building materials, including double-glazed panels and have real, broadly-understood meaning that is not just part of thermal conductivity. This is not a physics page, it warrants a full page of its own, and it should be left alone by the physicists. Jdpipe (talk) 00:06, 3 March 2009 (UTC)


 * Both the audience and the scope of the articles are different. R-Value (insulation) encompasses applied science and is primarily focused around the construction trade. The physics article is primarily theory and not taylored for a lay audience. I added the article: List of insulation material to redirect to the list in this article to make it easier to find. The only thing we'd accomplish by merging them would be to hide information about insulation for buildings, which would probably please the Arabs but would diminish the quality of Wikipedia. Helena srilowa (talk) 21:02, 24 March 2009 (UTC)


 * Perhaps we do our homework a little better. Joseph Fourier introduced the constant, H, relating the heat flux through the air to the temperature differences in The Analytical Theory of Heat. It is the present day U-value. He pointed out that it was different from the h used for thermal conductivity. Surely this takes priority over the HVAC definition. Fourier's discussion is quite modern and is similar to the derivation of Kirchhoff' law. He assumes there is a flow of heat in the absence of air. Where did the HVAC industry get its definition from? The article appears to have been stripped of its context. --Jbergquist (talk) 17:35, 12 July 2009 (UTC)


 * The Code of Federal Regulations indicates that the R-value is determined by heat flux measurements specified by ASTM procedures. Therefore, some information concerning the physical meaning of the R-value would be relevant. --Jbergquist (talk) 23:45, 12 July 2009 (UTC)


 * US Patent 5940784 - Heat flow meter instruments gives some details about the measurement process. Fourier's Law involving thermal conductivity is assumed and "a predicted steady-state value of a thermal property" is made. Q is the symbol used for heat flux, "heat flow per unit surface area", but to meet criticisms $$\dot Q_A$$ was used instead for added clarity in Fourier's formula for lamina. --Jbergquist (talk) 21:34, 1 August 2009 (UTC)

Definition of U-value
The concept of U-value is frequently first taught to engineers (at least mechanical engineers) in the context of heat exchangers, where it is introduced as the 'overall heat transfer coefficient' between one fluid, via a steel wall, to another fluid, including convection, conduction, convection. This definition is much more robust and universal, and needs to replace the definition given on this page. It should furthermore be pulled out into a page of its own, because most students will learn about U-values before they first hear about R-values and R_th. Jdpipe (talk) 01:52, 27 February 2009 (UTC)

Formula in lead
I mistakenly reverted a formula added to the lead:


 * It is the ratio of the temperature difference across an insulator and the heat flux through it, i.e., $$R={\Delta T}/{\phi_q}$$.

I was thinking of the math/electromagnetics definition of flux, not realizing that a different definition was used in transport phenomena. Sorry about that! Ccrrccrr (talk) 12:56, 13 July 2009 (UTC)


 * Thx. The general term might be energy flux and apparently that term also has some ambiguity in its use. There is a lot of variation in the symbols and terms used for "heat flux". A quick search of the internet showed that q and Q are often used. My edition of the Handbook of Chemistry and Physics lists a number terms used to express transport properties and they use Jq for heat flux (units W/m2). I used &phi;q to be consistent with the usage in heat flux and minimize the amount of confusion within Wikipedia. --Jbergquist (talk) 00:44, 14 July 2009 (UTC)


 * Minimizing the amount of confusion in WP sounds like a large project! More seriously, I do like the idea of Jq, because it would be more readily understandable to people who have a background in electromagnetics, since J is used for the analogous quantity current density.  Q is problematic because it's used both for quantity of heat and for heat flow.  Q dot would seem to me to indicate a time derivative of one of those, but is often used to indicate heat flow rather than heat--that seems problematic to me, though I see someone just put it in here that way.  (Derivative and flow are different concepts--it's quite possible to have steady non-zero flow with the derivatives all zero.)Ccrrccrr (talk) 01:20, 14 July 2009 (UTC)


 * The R-value relating the temperature difference and the heat flux is a formula for a transport phenomena. Consider the analogy with Darcy's law which relates the pressure difference across and the flow of water through some material and the similarity of the constant of proportionality with resistance. The formula we are working with is analogous with Ohm's law in which R is the constant of proportionality between the potential difference across and current through a resistor so making the changes you will overcomplicates matters. These analogies give us the Q and the R. I don't like the use of J because that is used for the current density vector and heat flux is a scalar. --Jbergquist (talk) 02:27, 14 July 2009 (UTC)


 * In general heat flux is a vector. See heat flux.  The thermal resistance equation, (heat flow, e.g. in watts) = &Delta;T/Rth, where Rth is thermal resistance in K/W, is analogous to ohm's law, not the equation with R-value in the position of R.  Ccrrccrr (talk) 02:47, 14 July 2009 (UTC)


 * The analogy isn't exact. What's involved in the analogy is a "potential difference", a scalar flow and a constant linking them. Ohm's law would involve "fluxes" (and an extra m2 unit) if we considered the flux of current through a square slab acting as the resistor. You don't need to worry about the thickness of the slab when "potential differences" are used. --Jbergquist (talk) 03:35, 14 July 2009 (UTC)


 * Here's a problem someone might like to add: What is the thermal resistance of 8 16"x96" sheets of R-19 insulator? At what rate will heat will flow through them if the temperature difference across them is 20 &deg;F? --Jbergquist (talk) 04:12, 14 July 2009 (UTC)

Would it help to remove some of the ambiguity in W/m2?
Using the watt to describe both the rate at which work is done and the rate of heat flow can be confusing in this situation. It may help if some derived unit could be used to measure the heat flux instead of W/m2. The best name for this unit of heat flux [ &phi;q ] may be the "fourier". It might be abbreviated as Fo avoiding confusion with the franklin, the Froude number or Francium and in conformity with the abbreviation for the Fourier number. Then the units for the R-value would then be [T]/[ &phi;q ] which would be &deg;C/Fo and that of the U-value would be Fo/&deg;C. --Jbergquist (talk) 22:07, 14 July 2009 (UTC)


 * Whether to use different unit names for energy flow in different situations is an interesting question, and one about which I'd love to tell you my strong opinions. However, it's not something that should be discussed on this page, because even if we editors developed a consensus on a great plan along those lines, we'd still be prohibited from mentioning it in this or any Wikipedia article, because it would be our original research.Ccrrccrr (talk) 00:58, 15 July 2009 (UTC)


 * I agree that the idea is somewhat premature and not suitable for inclusion in a Wikipedia article. But it occurred to me that we might encounter some absurd expression like W/W while discussing the rate of heat flow at a given power expenditure needed to maintain an equilibrium in temperatures where "Fo-m2/W" might be clearer. With respect to global warming passive systems may be preferrable and active systems counter-productive. High R-value is definitely low carbon but a building needs to allow for an exchange of gases too. --Jbergquist (talk) 05:31, 15 July 2009 (UTC)


 * "some absurd expression like W/W". Like Coefficient of Performance?Ccrrccrr (talk) 13:13, 15 July 2009 (UTC)

Use of non SI unit in preference to SI
The table detailingspecific R Values should be corrected to show SI values with regional and non SI units in brackets and not the other way around. — Preceding unsigned comment added by 86.1.65.206 (talk) 18:02, 20 September 2009 (UTC)


 * Agreed that table should be changed to SI units with non SI units in brackets. —Preceding unsigned comment added by 122.104.137.248 (talk) 11:39, 3 October 2009 (UTC)

Agreed, it seems that some of the "non-applicable", NA, North American, "scientists" wish to insert their own values into the international world wide SI units system without a full understanding of the basis of scientific talk. Please remove all "inches" and derivatives thereof from tables displaying values in SI units. kk (talk) 06:38, 10 June 2016 (UTC)


 * The non-SI values were put inside brackets some time ago. The current version of the table has them in a separate column altogether. kk, the reason for the inclusion of the "inch" with the SI units is given in "Incoherence in SI units" on this talk page. The R-value has its origin in the US. I don't think there's a conspiracy. Pololei (talk) 19:24, 17 June 2016 (UTC)

R-value for glass in table is incorrect
According to Halliday & Resnick, Fundamentals of Physics, Third Edition, 1988 on p. 473, R value of window glass in US units is 0.14. The table in the article shows 0.24. However if you start with the value on Wikipedia for the thermal conductivity of glass provided in the article on that topic, which is 1.1 in SI units, you can prove the 0.14 value is correct. R = thickness / thermal conductivity = 0.0254 m / 1.1 = 0.0231, which in US units is 0.1311.

Couldn't figure out how to edit the table myself, but the error just cost me several hours, so I thought I'd report it.

71.235.72.45 (talk) 03:32, 21 March 2010 (UTC)


 * Thank you for catching this! The reason why it was hard to edit was not (primarily) because it's in a table, but because the table has been transcluded, which is quite unusual. The actual page that needed to be changed was Insulation (list of insulation material). &mdash; Sebastian 22:45, 26 March 2010 (UTC)

Units conversion is not correct?
1 h·ft²·°F/Btu = 3600s·(0.3048m)²·1.8K/1055.05585J = 0.570597 K·m²/W. Not 0.176110 K·m²/W as in article.

No, the original (0.176110) is correct, see the ASHRAE Fundamentals reference. You multiplied by 1.8 °F/K, you should be dividing by 1.8 °F/K. Elfstrom (talk) 05:58, 17 October 2010 (UTC)

Typical value for drywall/plasterboard/gypsum board
Can we add to the table of typical values some of the more common construction materials eg drywall/plasterboard/gypsum board ? Rod57 (talk) 23:05, 30 November 2010 (UTC) Plywood would be useful too.
 * RSI 12.5mm drywall seems to be about 0.08. Rod57 (talk) 23:36, 30 November 2010 (UTC)

What thickness?
The article says that R-value is expressed as the thickness of the material divided by the thermal conductivity. The SI unit m2K/W is in accord with this, provided the thickness is included along with the thermal conductivity. Without a specified thickness the definition is meaningless.

The knee-jerk physicist response would be that what is really meant here is that [R-value] is simply the reciprocal of thermal conductivity, whose units would be mK/W. Otherwise, says the physicist, how could a quantity given in units of m2K/W be used? What if I have one sheet of glass of thickness 3 mm and another 6 mm? How do I use R-value consistently with its units to get different heat fluxes for each?

The only way I can make sense of this is if R-value is not a property of materials in the sense of physics but of construction materials of known thickness. You would then give separate R-values for 3 mm and 6 mm glass. Typically when selling a product by the square foot such as drywall or batting or window panes you would also have other attributes to hand, allowing other factors to be taken into account.

It seems to me that the article gets this wrong in the fourth sentence, This is used for a unit value of any particular material. A more accurate statement might be This is used for unit area of a construction material with known relevant attributes such as thickness, emissivity, etc.

As others have pointed out the temperature unit is wrong for radiation, in particular for emissivity, about which the article could presumably say something; it should be K4, not K. It could not of course propose unsourced solutions, but with the thought in mind that the building industry evidently refers to these talk pages, one solution might be to adopt a standard room temperature, say 20 &deg;C for SI units and 70 &deg;F for US, and give two R-values in units of m2/W or ft2h/Btu, one for each of two representative outside temperatures, from which the relative contributions of the conductive and radiative components might be deducible from the distance of the line connecting the two temperatures from the origin (outside temperature equal to inside). This would have the additional benefit of simplifying the units and using the value by removing temperature from both the unit and the computation. Convection is a bit of a spoiler there. --Vaughan Pratt (talk) 18:59, 15 December 2010 (UTC)


 * Seems that no-one answered this for some years, but yes it is for specific thicknesses. In many materials, it is a combination of conduction and (hopefully slow) convection.  Consider popular glass fiber insulation as an example. You buy a batt or roll of such, with a design thickness.  It is, then, not necessarily proportional to thickness.  If there is an air flow barrier, then you can add the values of stacked insulation.  If there isn't, it might be a little different.  Also, compressing it can change the R value. Gah4 (talk) 04:36, 11 February 2019 (UTC)

Linear relationship
There is a "citation needed" annotation, which I think is intended to express a need for documentation or adjustment to the phrase "this linear relationship may be only approximate for some materials".

While this might not be the best phrasing, I think the original purpose of the phrase was to indicate that the R-value of a structure cannot necessarily be determined by multiplying the "R-value per inch" of its component material by its thickness. A clear example is an airspace; the R-value of an 1/2 inch airspace is about 1 (in US units), while the R-value of a 4 inch space is virtually the same.

The FTC determined that the potential for misleading claims was so high they essential banned the use of "R-value per inch" as an advertising claim.

I'm not so clear on the best way to present this in the context of the article. — Preceding unsigned comment added by Wcoole (talk • contribs) 22:56, 10 February 2011 (UTC)


 * Also, if you compress some materials, such as glass fiber, the value will change in a not so obvious way. Gah4 (talk) 04:39, 11 February 2019 (UTC)

Straw Bale values
The cited values for straw bale R the value is with respect to a study that even the link referenced said it is questionable (circulation behind drywall facing changing value). As such, and since this page is talking about basic R-values and not combinational values, shouldn't the more accepted values be used?

eg R-2.68 per inch to R-2.38 per inch, rather than R-0.94 per inch SanityChek (talk) 03:08, 15 November 2011 (UTC)

Removed comparison to concrete
I remove the following phrase: "The general rule for the general effectiveness of insulation is that one inch of insulation roughly equals 30 inches of concrete.", along with reference to http://rvalue.net/ as it is impossible to create a direct comparison between "insulation" and "concrete." The link simply states this as fact. This is similar to saying "buses" carry 60 times more passengers than "cars" - simply too vague to be useful. Andy (talk) 10:13, 6 February 2012 (UTC)

R-factor (crystallography)
Could some one add "This article is about X, look R-factor (crystallography) for Y" template to article? &mdash; linnea (talk) 17:28, 10 May 2012 (UTC)

Those of us watching this page might not be qualified to write the "Y" part of that. (at least I'm not). Can you suggest text for that, or perhaps post a comment on that page to request that help here? Ccrrccrr (talk) 21:37, 10 May 2012 (UTC)


 * "This article is about R-value in thermal resistance. Look R-factor (crystallography) for R-value used in chrystallography." &mdash; linnea (talk) 08:15, 11 May 2012 (UTC)

U-value discussion

 * The article is R-value and the discussion is U value. Let's have an R-value discussion here and a U value discussion elsewhere.

Dhunt16 (talk) 17:05, 29 November 2014 (UTC)


 * Since U is just 1/R, there doesn't seem to be a lot of reason for a separate article, but one could be added. R is convenient as it is close to additive. For now, U-value redirects here, so it should be described here. Gah4 (talk) 04:46, 11 February 2019 (UTC)

Adding other materials to the "Example values" list
I request adding cork to the "Example values" list. I am not proficient at Wikipedia editing or insulation, otherwise I would do it myself. — Preceding unsigned comment added by Jdmumma (talk • contribs) 18:18, 23 March 2015 (UTC)


 * Some materials are variable. Though they could be given as a range. Gah4 (talk) 04:47, 11 February 2019 (UTC)
 * Yes, but not clear on why you focus on "range" ("could be given as a range")
 * 1. "range" is already noted "Typical R-values are given for various materials and structures as approximations based on the average of available figures and are sorted by lowest value. R-value at 1 m gives R-values normalised to a 1 metre (3 ft 3 in) thickness and sorts by median value of the range."
 * 2. "Vacuum insulated panel" is already listed as a range of: 5.28–8.8 / 14–66
 * 3. Looks like ~30% of the material listed have a range Jdmumma (talk) 16:15, 23 January 2023 (UTC)

Removed from "aging" subsection
"There has been a test method conceived to test the flammability of thermal/acoustic insulation. This type of insulation usually contains a thin film of moisture barrier over a batting material, with the possibility of foam being a second barrier. The test also takes into account small detail parts of the insulation which might contribute to whether or not the insulation is flammable. Such details include thread, tape, and fasteners. The test consists of putting the insulation next to an ignition source, then observing whether or not it catches fire. Then, if the specimen has caught fire, the ignition source is removed and the insulation is observed to see if it continues to burn. "

This has nothing to do with R-value; it also made no sense where it was in the article. Huw Powell (talk) 06:27, 9 November 2015 (UTC)

Practical meaning of U-value
The article states that the higher the R Value, the better the insulation effect. So given that U Value is the inverse of R Value, then the lower the U Value, the better the insulation effect. It would be nice for the article to say so. 173.79.59.113 (talk) 16:24, 21 January 2016 (UTC)


 * The section on U-value says, "A low U value usually indicates high levels of insulation". It is implied in the lead section, too. Pololei (talk) 18:39, 21 January 2016 (UTC)

R values of Hempcrete
I was looking for R values for hempcrete. Closest thing I found was a .AU site saying "R4 for 300mm". This may be US or SI units - probably the latter. I also didn't see any way to add the entry to the table. — Preceding unsigned comment added by 70.66.191.220 (talk) 01:50, 23 November 2016 (UTC)

Use of Kelvin
The use of Kelvin when a temperature difference seems to make the article less accessible. Most normal people think of temperature in Celsius and examples give a temperature in Celsius. As the 'RSI' value is a delta of temperature the 'C°' notation can be used. Especially as 'K' can be confused with a 'Kilo' notation. Many people will not be familiar with Kelvin and having random 'K' as an abbreviation makes the article less accessible even if it is the unit used by technical people. 108.161.125.86 (talk) 13:43, 17 April 2017 (UTC)


 * Although it's less widely recognised than the degree Celsius, the kelvin is the SI unit for temperature. RSI is numerically equivalent to thermal insulance when expressed in SI units. Pololei (talk) 04:25, 31 March 2019 (UTC)

Too myopic by half
However, the LTTR effectively provides only an eight-year aged R-value, short in the scale of a building that may have a lifespan of 50 to 100 years.

But not short in net present value calculations. This sentence is written as if the time discount is non-existent on economic napkins. Furthermore, the sentence is rife with unspecified assumptions about the future cost of energy, both as direct cost and in terms of social externalities (which some home builders file under the myth of precipitous, adverse climate change). &mdash; MaxEnt 18:08, 24 September 2017 (UTC)
 * Only to provide a contrasting opinion to support the original wording: directed towards someone who is interested in evaluating the ability of a 60 year-old building to retain heat, the lack of data on insulating materials over longer periods of time makes objective analysis difficult. But I agree the wording could be worded better, how about: "The LTTR effectivley provides an eight-year aged R-value for insulative materials, but cannot be used to estimate the degradation of R-values over longer periods of time."

External links modified
Hello fellow Wikipedians,

I have just modified 4 external links on R-value (insulation). Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20070112202600/http://www.ornl.gov/sci/roofs+walls/insulation/ins_02.html to http://www.ornl.gov/sci/roofs+walls/insulation/ins_02.html
 * Added archive https://web.archive.org/web/20070112202600/http://www.ornl.gov/sci/roofs+walls/insulation/ins_02.html to http://www.ornl.gov/sci/roofs+walls/insulation/ins_02.html
 * Added archive https://web.archive.org/web/20120609025242/http://www.professionalroofing.net/archives/past/feb02/polyisofeature.asp to http://www.professionalroofing.net/archives/past/feb02/polyisofeature.asp
 * Added archive https://web.archive.org/web/20050405145911/http://www.plastics.org.nz/page.asp?id=468 to http://www.plastics.org.nz/page.asp?id=468

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 20:09, 9 December 2017 (UTC)

pipe insulation
I was looking up the pipe insulation, which has cylindrical symmetry, and which is given an R value. I understand how resistivity works in radial (cylindrical) coordinates, but it isn't so obvious here. This article assumes flat materials, with the same area on each side, which is not true for pipe insulation. Gah4 (talk) 11:11, 22 January 2019 (UTC)


 * There's an article on pipe insulation that has a section on what you are interested in: Pipe_insulation Someone should link it in the article. Ccrrccrr (talk) 02:59, 11 February 2019 (UTC)


 * Interesting, and conveniently doesn't answer the question. After the post, I was thinking that it should be per linear (distance), since (as with flat insulation) it is rated based on the specific dimensions. (That is, different diameters are quoted separately.)  But I thought it could also be based on either the inside our outside area, which the one you point to conveniently ignores.  Thanks, though. Gah4 (talk) 04:15, 11 February 2019 (UTC)

sig figs
Do the references, which I don't have available, really give the conversion factors to/from SI units, to 10 significant digits? Considering that insulation materials are probably accurate to one digit, that seems excessive. Gah4 (talk) 03:32, 9 February 2019 (UTC)


 * I agree, that's excessive, but technically it's probably valid, given that that's derived from other unit conversions like inches to meters and joules to BTU, which are defined very precisely.Ccrrccrr (talk) 03:01, 11 February 2019 (UTC)


 * If the reference has the digits, then it is hard to complain, even if excessive. Recomputing the constant from known conversions is, as you note, technically valid, but I think not necessary. I suspect that there are no cases when one needs that conversion, even more generally than R values, to that many digits.  I was in college not long after calculators pretty well replaced slide rules, and giving values to 10 digits was common, and TAs were good at marking them off.  I remember once when the answer involved only the constants c, e, and 2, where the 2 should be exact.  I was still marked off for excessive digits.  I don't have the reference available to look them up, though. Gah4 (talk) 04:06, 11 February 2019 (UTC)

"two-dimensional barrier"
I understand what the writer was intending to convey with this. However, two-dimensional is not accurate in this case, as some degree of thickness is always required. — Preceding unsigned comment added by 206.174.219.186 (talk) 19:08, 21 January 2020 (UTC)

"Orders of magnitude (thermal insulance)" listed at Redirects for discussion
An editor has asked for a discussion to address the redirect Orders of magnitude (thermal insulance). Please participate in the redirect discussion if you wish to do so. Utopes (talk / cont) 22:24, 11 April 2020 (UTC)

"Square metre-Kelvin per Watt" listed at Redirects for discussion
An editor has asked for a discussion to address the redirect Square metre-Kelvin per Watt. Please participate in the redirect discussion if you wish to do so. Utopes (talk / cont) 22:24, 11 April 2020 (UTC)

Source for Four Materials in Aging of Insulative Materials
I found this article on the effects of aging on the R-value of some insulative materials. I haven't been able to check out the paper itself, but I would think that the National Center for Preservation Technology and Training is a trust-worthy source on this topic. If anyone can add some of this info to the topic on the effect of aging, I think that would be great. --IanVG (talk) 19:31, 27 September 2021 (UTC)
 * Actually, I went ahead and made the changes I was considering. Please let me know if they seem appropriate. I think the section on aging could use an hour or two of dedicated time looking up trustworthy sources on this topic and adding summarized version (or simply their existence as references) to add more credence to the article's coverage. --IanVG (talk) 19:48, 27 September 2021 (UTC)

Partial Deletion of Introduction
"U-Values of materials are found by observation under standardised conditions. The usual standard is at a temperature difference of 24 °C, at 50% humidity with no wind (a smaller U-factor is better at reducing heat transfer). It is expressed in watts per square metre kelvin (W/m2⋅K)."

This seems out of place and the link does direct to a valid source anymore. Also, I may be wrong but I thought I remember reading that the R-value (and thus the U-Value) for commercially sold products is calculated by using a mean temperature of 75F with a base of 50F and a max of 100F. This may be a ASTM or ASHRAE test (ie. american-centric), so maybe acknowledging that other standards are present would be a good idea. I also don't think the lengthy discussion about the U-value is appropriate here. --IanVG (talk) 20:10, 27 September 2021 (UTC)

The chart is full of problems
The chart, Typical R-values, is full of calculation errors. I'm not sure how these were sourced, but when I started plugging values into a spreadsheet to get a ratio, I ran into some oddities. For instance, the ratio between "RSI-value (m2·K/W)" and "RSI-value (m2·K/W) (per metre)", appears to be about 0.025, but for Insulated Glass, the ratio reported is 0.00875. I was thinking about trying to fill in a couple holes, but if the numbers are as trustworthy as they appear, I'm not sure I should bother.

I'd just go in and start changing things, but some of them have citations. Sean.Roach (talk) 22:48, 21 June 2022 (UTC)