# ASTM C653-17

Designation: C653 − 17Standard Guide forDetermination of the Thermal Resistance of Low-DensityBlanket-Type Mineral Fiber Insulation1This standard is issued under the fixed designation C653; the number immediately following the designation indicates the year oforiginal adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. Asuperscript epsilon (´) indicates an editorial change since the last revision or reapproval.1. Scope1.1 This guide describes the calculation and interpolation ofa thermal resistance value for low-density blanket-type insula-tion material at a particular density and thickness having beenselected as representative of the product. It requires measuredvalues of this average density and thickness, as well asapparent thermal conductivity values determined by either TestMethod C177, C518,orC1114.1.2 This guide applies to a density range for mineral-fibermaterial of roughly 6.4 to 48 kg/m3(0.4 to 3.0 lb/ft3). It isprimarily intended to apply to low-density, mineral-fiber massinsulation batts and blankets, exclusive of any membranefacings. Apparent thermal conductivity data for these productsare commonly reported at a mean temperature of 23.9°C (75°F)and a hot-to-cold plate temperature difference of 27.8°C (50°F)or 22.2°C (40°F).1.3 This standard does not purport to address all of thesafety concerns, if any, associated with its use. It is theresponsibility of the user of this standard to establish appro-priate safety and health practices and determine the applica-bility of regulatory limitations prior to use.2. Referenced Documents2.1 ASTM Standards:2C167 Test Methods for Thickness and Density of Blanket orBatt Thermal InsulationsC168 Terminology Relating to Thermal InsulationC177 Test Method for Steady-State Heat Flux Measure-ments and Thermal Transmission Properties by Means ofthe Guarded-Hot-Plate ApparatusC518 Test Method for Steady-State Thermal TransmissionProperties by Means of the Heat Flow Meter ApparatusC687 Practice for Determination of Thermal Resistance ofLoose-Fill Building InsulationC1045 Practice for Calculating Thermal Transmission Prop-erties Under Steady-State ConditionsC1114 Test Method for Steady-State Thermal TransmissionProperties by Means of the Thin-Heater Apparatus3. Terminology3.1 Definitions—For definitions used in this guide, refer toTerminology C168.3.2 Definitions of Terms Specific to This Standard:3.2.1 apparent thermal conductivity, λ—the ratio of thespecimen thickness to thermal resistance of the specimen. It iscalculated as follows:λ 5 L/R ~W/m·k! or ~Btu·in./ft2·h·F! (1)3.2.1.1 Discussion—For this type of material an expressionfor the apparent thermal conductivity as a function of densityis:λ 5 a1bD1c/D (2)where a, b, c = parameters characteristic of a product,and related to the conductivity of the gas, the conductivityof the solid and the conductivity due to radiation (1).33.3 Symbols:R = thermal resistance, (m2K/W) or (h·ft2F/Btu)λ = apparent thermal conductivity, (W/m·K) or (Btu·in/h·ft2F)Q/A = heat flow per unit area, (W/m2) or (Btu/h·ft2)D = bulk density of a specimen, (kg/m3) or (lb/ft3)L = measured specimen thickness, (m) or (in.)T = apparatus plate temperature, (K) or (F)L = specimen thickness if the sample from which thespecimen is selected does not recover to labelthickness, (m) or (in.)s = estimate of the standard deviation for a set of datapoints∆ = apparatus systematic errorΨ = overall uncertainty in a measured R-value1This guide is under the jurisdiction of ASTM Committee C16 on ThermalInsulation and is the direct responsibility of Subcommittee C16.30 on ThermalMeasurement.Current edition approved March 1, 2017. Published March 2017. Originallyapproved in 1970. Last previous edition approved in 2012 as C653 – 97 (2012).DOI: 10.1520/C0653-17.2For referenced ASTM standards, visit the ASTM website, www.astm.org, orcontact ASTM Customer Service at service@astm.org. For Annual Book of ASTMStandards volume information, refer to the standard’s Document Summary page onthe ASTM website.3The boldface numbers in parentheses refer to a list of references at the end ofthis standard.Copyright © ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. United StatesThis international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for theDevelopment of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.13.3.1 Subscripts:av= signifies average of a lotH= refers to hot surfaceC= refers to cold surfaceT= refers to test specimenN= refers to nominal property for the product, as shown onthe product labeli= refers to a set of data pointss= refers to a particular specimen4. Significance and Use4.1 This guide provides a method to determine the thermalperformance of low-density blanket-type insulation. It may beused for the purposes of quality assurance, certification, orresearch.4.2 The thermal resistance of low-density insulation de-pends significantly on the density, the thickness, and thermalconductivity. Typical low-density, mineral-fiber insulation forbuildings may vary in density from one specimen to the next.4.3 Thermal tests are time-consuming in comparison withdensity and thickness measurements. Low-density insulationmaterial is produced in large quantities. A typical lot would bea truckload or the amount necessary to insulate a house.4.4 The relatively low unit cost of this product and therelatively high cost of thermal resistance testing makes itcost-effective to test only a small percentage of the productarea. It is recommended that there be a determination of thedensity that is representative of a lot by the measurement of theaverage density of a statistically representative sampling.4.5 Afewer number of thermal measurements are then madeto determine the apparent thermal conductivity at the previ-ously determined representative density. The essential signifi-cance of this guide is that a large lot of variable material is bestcharacterized by: (a) determining the representative density,and by (b) determining the thermal property at this represen-tative density with a small number of thermal measurements.4.6 Building insulation products are commonly manufac-tured in thicknesses ranging from 19 to 330 mm (0.75 to 13 in.)inclusive. Experimental work has verified that there is adependence of λappon thickness for some low density materi-als.4.7 The upper limit of test thickness for specimens evalu-ated using Test Methods C177, C518, and C1114 is establishedbased upon the apparatus design, overall dimensions, expectedthermal resistivity level and desired target accuracy. Thetesting organization is responsible for applying these restric-tions when evaluating a product to ensure that the results meetapplicable product labels and any existing regulatory require-ments (2).4.8 Extrapolation of the apparent thermal conductivity orthe thermal resistance beyond the ranges of thickness ordensity of products tested is not valid.5. Sampling5.1 For low-density mineral-fiber insulation, a lot samplesize of 75 to 150 ft2is recommended to determine the averagedensity, Dav. Density is determined by using Test MethodC167; take care to avoid the use of damaged material.5.2 In order to account for the variation in λ-value due toproduct density variability, measure a minimum of three “λversus D” data points on three different samples. This repre-sents nine data points for the “λ versus D” curve.Again, this “λversus D” curve is developed to determine the λ-value at aparticular representative density characteristic of a lot ofmaterial.5.3 The size of a lot of material to be characterized, theamount of material measured for the representative values ofdensity and thickness, and the frequency of tests all depend onthe user’s needs, which could be related to quality assurance bya manufacturer, certification, or research.6. Procedure6.1 This procedure uses nine {λi; Di} data points allmeasured at the same hot and cold plate temperatures, toestablish an interpolation equation for the determination of theλ-value at the average density, Dav. That is, the subscript irefers to the ithtest point. The DIis the average density of thespecimen within the apparatus meter-area. The thermal resis-tance at Lavand Davis as follows:Rav5 Lav/λav(3)6.2 Before the set of “apparent thermal conductivity versustest density (λiversus Di)” data points can be measured on anapparatus, it is necessary to choose the test densities andthicknesses. Three procedures for this choice are described inAnnex A1.6.2.1 Procedure A—Asingle test specimen is compressed toobtain different densities (A1.2). This procedure offers theadvantage of less test time to obtain three test points.6.2.2 Procedure B—A different specimen is used for eachtest point (A1.3). This method has the advantage of a betterstatistical sampling with regard to material variability.6.2.3 Procedure C—Test at Davthereby eliminating the needfor an interpolation (A1.4).6.3 Obtain a test value for λ at each of the three densities.These three sets of test values result in three equations of theform of Eq 2 in 3.2.1. These are solved simultaneously todetermine the values of as, bs, and cscorresponding tospecimen s (see A2.1.2).NOTE 1—Small errors in the measured values of λ will result in largevariations in the values of a, b, and c. Even so, the uncertainty of theinterpolated value of λ will be comparable to the measured error in λ.6.4 Whenever possible, calculate running averages for thespecific product lot based on a number N equal to 20 or moresets of product curve parameters (as;bs;cs). Remember from6.3 that each of these sets requires three test points (seeA2.1.3).6.4.1 A larger number N results in more consistent valuesfor a, b, and c; a smaller N represents a more current data base.6.5 In 6.3 a set of parameter values was calculated, and in6.4 a running average was calculated. This section describeshow to obtain an interpolation curve (or equivalently a set ofinterpolation curve parameters) for the next sample, s, when itC653 − 172has been possible to previously obtain a running average set,(a¯; b¯; c¯). The given values are the set {a¯; b¯; c¯} and themeasured values of λiat three densities, Di.NOTE 2—Parameter c is expected to account for most of the variation inthe “λ versus D” curve from specimen to specimen. When the density isless than 16 kg/m3(1 lb/ft3), c is the dominant parameter causing thevariance of λ from specimen to specimen. Then the previously determinedvalues, a¯, and b are used, along with a measurement of λ at a particulardensity, to calculate a value of c for a particular specimen, s. In order tohave a better estimate of the mean, the value of c is thusly determined forthree values of density resulting in the value c¯s. The interpolation to the λvalue at the average density, Dav, is calculated as follows, using Eq 3.λs5 a¯ 1b¯Dav1c¯s/Dav(4)An example of this calculation is in A2.1.46.6 Compute the average value of λ¯avbased on as manyvalues of λsthat have been determined. Remember from 6.3and 6.5 that three test points are required to obtain a value forλav. Common practice is to base an average λ¯avon three valuesof λs.6.7 Calculate the R-value, Rav, of the product at the averagedensity and thickness (see Section 5 and A1.1) as follows:Rav5 LT/λav(5)7. Report7.1 The report shall contain the following information:7.1.1 The values of the average thermal resistance, densityand thickness, the sample size, and the supporting data.7.1.2 The test methods used and the information on thevalues and uncertainties of apparent thermal conductivity anddensity that is required in Test Method C167, C177, C518,orC1114.7.1.3 The procedure used to obtain the λ versus D curvealong with the equation for the curve itself.8. Precision and Bias8.1 There are a number of ways to combine the systematicand random uncertainties that contribute to an overall uncer-tainty of a measured quantity. The following procedure isintended as a guideline.8.2 The term precision is used in this guide in the sense ofrepeatability. The estimation of the standard deviation, s, for aset of measurements with a normal distribution is the plus andminus range about an average value or curve, within which68 % of the observations lie. The s is used to quantify theprecision.8.3 The term bias as used in this guide represents the totaluncertainty in a set of measurements, including apparatussystematic error, apparatus precision, and the material variabil-ity.8.4 The apparatus precision is the variation that occurswhen repeated observations are made on a single specimen oridentical specimens. It is quantified by sa, and it is required asinput data from either Test Method C177, C518,orC1114 (3).8.5 The material variability is partly taken into account bythe λ versus D curve. When different specimens are tested therewill be an amount of variation about the average λ versus Dcurve in addition to the apparatus precision. This additionalvariation is here called the material variability and is desig-nated by sm.8.6 The total “repeatability” uncertainty on a λ versus Dgraph will be the sum of the aforementioned uncertainties andis designated by sλ.sλ5 ~sa21sm2!0.5(6)8.7 In order to know what sλis, it is necessary to plot anumber of λ versus D test points. Twenty or more points arerecommended. It is then possible to determine by a graphical ora mathematical method (see Annex A3) what is the 1s bandwithin which 68 % of the points lie or what is the 2s bandwithin which 95 % of the points lie.8.8 When more than one apparatus is used to develop the λversus D curve, there will be a difference between the averagevalues on the same set of specimens due to a systematicdifference among the apparatus.8.9 The measured data from an apparatus have associatedwith it an estimate of the possible systematic error in λ of thatapparatus. It is designated by ∆λand is provided as input fromTest Method C177, C518,orC1114.8.10 For the purposes of this guide the overall accuracy, Ψλ,of the reported λ-value is the sum of the overall repeatability(1s for a 68 % confidence band) and the apparatus systematicerror.Ψλ5 sλ1∆λ(7)8.11 The percent “precision and bias” uncertainties in thereported R-value is calculated as follows, based on Eq 1:Rav5 LT/λav(8)8.11.1 The estimate of the residual standard deviation of Lavand λavis made by statistical methods (see Annex A3). Thepercent residual standard deviation in the reported R-value isthen:sRRav5SsL2LT21sλ2λr2 D0.5(9)8.11.2 In order to calculate the percent bias uncertainty inRv, it is necessary to obtain from Test Method C167 theestimate of systematic uncertainty in the measurement of Lav.This is of the order of the resolution of the measurementdevice, and it is designated here by ∆L. For the purpose of thisguide,