1、Designation: D 6600 00 (Reapproved 2004)Standard Practice forEvaluating Test Sensitivity for Rubber Test Methods1This standard is issued under the fixed designation D 6600; the number immediately following the designation indicates the year oforiginal adoption or, in the case of revision, the year o
2、f last revision. A number in parentheses indicates the year of last reapproval. Asuperscript epsilon (e) indicates an editorial change since the last revision or reapproval.1. Scope1.1 This practice covers testing to evaluate chemical con-stituents, chemical and physical properties of compoundingmat
3、erials, and compounded and cured rubbers, which mayfrequently be conducted by one or more test methods. Whenmore than one test method is available, two questions arise:Which test method has the better (or best) response to ordiscrimination for the underlying fundamental property beingevaluated? and
4、Which test method has the least error? Thesetwo characteristics collectively determine one type of technicalmerit of test methods that may be designated as test sensitivity.1.2 Although a comprehensive and detailed treatment, asgiven by this practice, is required for a full appreciation of testsensi
5、tivity, a simplified conceptual definition may be givenhere. Test sensitivity is the ratio of discrimination power for thefundamental property evaluated to the measurement error oruncertainty, expressed as a standard deviation. The greater thediscriminating power and the lower the test error, the be
6、tter isthe test sensitivity. Borrowing from the terminology in elec-tronics, this ratio has frequently been called the signal-to-noiseratio; the signal corresponding to the discrimination power andthe noise corresponding to the test measurement error. There-fore, this practice describes how test sen
7、sitivity, genericallydefined as the signal-to-noise ratio, may be evaluated for testmethods used in the rubber manufacturing industry, whichmeasure typical physical and chemical properties, with excep-tions as noted in 1.3.1.3 This practice does not address the topic of sensitivity forthreshold limi
8、ts or minimum detection limits (MDL) in suchapplications as (1) the effect of intentional variations ofcompounding materials on measured compound properties or(2) the evaluation of low or trace constituent levels. Minimumdetection limits are the subject of separate standards.1.4 This standard does n
9、ot 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.1.5 The content of this practice is as
10、follows:SectionScope 1Referenced Documents 2Terminology 3Summary of Practice 4Significance and Use 5Measurement Process 6Development of Test Sensitivity Concepts(Absolute and Relative Test Sensitivity, Limited and ExtendedRange Test Sensitivity, Uniform and Nonuniform Test Sensitivity)7Steps in Cond
11、ucting a Test Sensitivity Evaluation Program 8Report for Test Sensitivity Evaluation 9Keywords 10Annex A1Background on: Use of Linear Regression Analysis andPrecision of Test Sensitivity EvaluationAppendix X1Two Examples of Relative Test Sensitivity Evaluation:Relative Test Sensitivity: Limited Rang
12、eThree ProcessabilityTestsRelative Test Sensitivity: Extended RangeCompliance versusModulusAppendix X2Background on: Transformation of Scale and Deriva-tion of Absolute Sensitivity for a Simple Analytical Test2. Referenced Documents2.1 ASTM Standards:2D 4483 Practice for Evaluating Precision for Tes
13、t MethodStandards in the Rubber and Carbon Black ManufacturingIndustries3. Terminology3.1 A number of specialized terms or definitions are re-quired for this practice. They are defined in a systematic orsequential order from simple terms to complex terms; thesimple terms may be used in the definitio
14、n of the more complexterms. This approach generates the most succinct and unam-biguous definitions. Therefore, the definitions do not appear inthe usual alphabetical sequence.3.2 Definitions:3.2.1 calibration material, CM, na material (or otherobject) selected to serve as a standard or benchmark ref
15、erencematerial, with a fully documented FP reference value for a testmethod; the calibration material, along with several othersimilar materials with documented or FPreference values, may1This practice is under the jurisdiction ofASTM Committee D11 on Rubber andis the direct responsibility of Subcom
16、mittee D11.16 on Application of StatisticalMethods.Current edition approved Dec. 1, 2004. Published December 2004. Originallyapproved in 2000. Last previous edition approved in 2000 as D 6600 - 00e2.2For referenced ASTM standards, visit the ASTM website, www.astm.org, orcontact ASTM Customer Service
17、 at serviceastm.org. For Annual Book of ASTMStandards volume information, refer to the standards Document Summary page onthe ASTM website.1Copyright ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.be used to calibrate a particular test method o
18、r may be used toevaluate test sensitivity.3.2.1.1 DiscussionAfully documented FP or FP referencevalue implies that an equally documented measured propertyvalue may be obtained from a MP = f (FP) relationship.However, unless f = 1, the numerical values for the MP and theFP are not equal for any CM.3.
19、2.2 fundamental property, FP, nthe inherent or basicproperty (or constituent) that a test method is intended toevaluate.3.2.3 measured property, MP, nthe property that themeasuring instrument responds to; it is related to the FP by afunctional relationship, MP = f (FP), that is known or that maybe r
20、eadily evaluated by experiment.3.2.4 reference material, RM, na material (or other ob-ject) selected to serve as a common standard or benchmark forMP measurements for two or more test methods; the expectedmeasurement value for each of the test methods, designated asthe reference value, may be known
21、(from other sources) or itmay be unknown.3.2.5 testing domain, nthe operational conditions underwhich a test is conducted; it includes description of the testsample or specimen preparation, the instrument(s) used (cali-bration, adjustments, settings), the selected test technicians,and the surroundin
22、g environment.3.2.5.1 local testing, na testing domain comprised of onelocation or laboratory as typically used for quality control andinternal development or evaluation programs.3.2.5.2 global testing, na testing domain that encom-passes two or more locations or laboratories, domestic orinternation
23、al, typically used for producer-user testing, productacceptance, and interlaboratory test programs.3.2.6 Although a simplified conceptual definition of testsensitivity was given in the Scope, a more detailed but stillgeneral definition using quantitative terms is helpful forpreliminary discussion.3.
24、2.6.1 test sensitivity (generic), na derived quantity thatindicates the level of technical merit of a test method; it is theratio of the test discrimination power or signal, that is themagnitude of the change in the MP for some unit change in therelated FP of interest, to the noise or standard devia
25、tion of theMP.3.2.6.2 DiscussionThis definition strictly applies to anabsolute sensitivity, see 7.2. The change in the FP may be anactual measurement unit or a selected FP difference. Therelation between the MP and the FP is of the form MP = f (FP).4. Summary of Practice4.1 This practice develops th
26、e necessary terminology andthe required concepts for defining and evaluating test sensitiv-ity for test methods. Sufficient background information ispresented to place the standard on a firm conceptual andmathematical foundation. This allows for its broad applicationacross both chemical and physical
27、 testing domains. The devel-opment of this practice draws heavily on the approach andtechniques as given in the referenced literature.3,44.2 After the introduction of some general definitions, abrief review of the measurement process is presented, suc-ceeded by a development of the basic test sensit
28、ivity concepts.This is followed by defining two test sensitivity classifications,absolute and relative test sensitivity and two categories, (1) fora limited measured property range and (2) for an extendedproperty range evaluation. For an extended property range foreither classification, two types of
29、 test sensitivity may exist, (1)uniform or equal sensitivity across a range of properties or (2)nonuniform sensitivity which depends on the value of themeasured properties across the selected range.4.3 Annex A1 is an important part of this practice. Itpresents recommendations for using linear regres
30、sion analysisfor test sensitivity evaluation and recommendations for evalu-ating the precision of test sensitivity.4.4 Appendix X1 is also an important adjunct to thispractice. It gives two examples of relative test sensitivitycalculations: (1) for a limited range or spot check program and(2) for an
31、 extended range test sensitivity program with adependent (nonuniform) test sensitivity. Appendix X2 givesbackground on transformation of scale often needed for ex-tended range sensitivity and for improved understanding, italso gives the derivation of the absolute test sensitivity for asimple analyti
32、cal chemical test.5. Significance and Use5.1 Testing is conducted to make technical decisions onmaterials, processes, and products. With the continued growthin the available test methods for evaluating scientific andtechnical properties, a quantitative approach is needed to selecttest methods that h
33、ave high (or highest) quality or technicalmerit. The procedures as defined in this practice may be usedfor this purpose to make testing as cost effective as possible.5.2 One index of test method technical merit and impliedsensitivity frequently used in the past has been test methodprecision. The pre
34、cision is usually expressed as some multipleof the test measurement standard deviation for a defined testingdomain. Although precision is a required quantity for testsensitivity, it is an incomplete characteristic (only one half ofthe necessary information) since it does not consider thediscriminati
35、on power for the FP (or constituent) being evalu-ated.5.3 Any attempt to evaluate relative test sensitivity for twodifferent test methods on the basis of test measurementstandard deviation ratios or variance ratios, which lack anydiscrimination power information content, constitutes an in-valid quan
36、titative basis for sensitivity, or technical meritevaluation. Coefficient of variation ratios (which are normal-ized to the mean) may constitute a valid test sensitivity3Mandel, J., and Stiehler, R.D., Journal of Research of National Bureau ofStandards, Vol 53, No. 3, September 1954. See also “Preci
37、sion Measurement andCalibrationStatistical Concepts and Procedures,” Special Publication 300, Vol 1,National Bureau of Standards, 1969, pp. 179155). (The National Bureau ofStandards is now the National Institute for Standards and Technology.)4“The Statistical Analysis of Experimental Data,” Chapters
38、 13 and 14, J.Mandel, Interscience Publishers (John Wiley for every value of MP there must be aunique single value for FP. The relationship must be specificfor any particular measuring process or test, and, if there aretwo different processes or tests for evaluating the FP, therelationship is genera
39、lly different for each test.7. Development of Test Sensitivity Concepts7.1 Test DomainThe scope of any potential test sensitivityevaluation program should be established. Is the evaluation fora limited local testing situation, that is, one laboratory or testlocation? Or are the results to be applied
40、 on a global basisacross numerous domestic or worldwide laboratories or loca-tions? If local testing is the issue, the test measurements areconducted in one laboratory or location. For global testing, aninterlaboratory test program (ITP) must be conducted. Two ormore replicate test sensitivity evalu
41、ations are conducted ineach participating laboratory and an overall or average testsensitivity is obtained across all laboratories. In the context ofan ITP for global evaluation, each replicate sensitivity evalu-ation is defined as the entire set of operations that is requiredto calculate one estima
42、ted value for the test sensitivity. Foradditional background on the assessment of precision for thetest sensitivity values attained, see A1.2 and also PracticeD 4483.7.2 Test Sensitivity ClassificationThere are two classifi-cations for test sensitivity.7.2.1 Class 1 is absolute test sensitivity, or
43、ATS, where theword absolute is used in the sense that the measured propertycan be related to the FP by a relationship that gives absolute ordirect values for FP from a knowledge of the MP. In evaluatingtest sensitivity for this class, two or more CMs are used eachhaving documented values for the FP.
44、7.2.2 Class 2 is relative test sensitivity, or RTS, where thetest sensitivity of Test Method 1 is compared to Test Method 2,on the basis of a ratio, using two or more RMs with differentMP values. This class is used for physical test methods whereno FPs can be evaluated.7.3 Absolute Test SensitivityI
45、n this section absolute ordirect test sensitivity is defined in a simplified manner by theuse of Fig. 1.7.3.1 Development of Absolute Test SensitivityFig. 1 isconcerned with two types of properties: (1) an FP (or singlecriterion or constituent), the value of which is established bythe use of a CM an
46、d (2) an MP obtained by applying the testmethod to the CM. A relationship or functionality existsbetween the MP and FP that may be nonlinear. In theapplication of a particular test, FP1corresponds to MP1andFP2corresponds to MP2. Over a selected region of therelationship, designed by points a and b,
47、the slope, K, of theillustrated curve is approximated by the relationship K =D(MP)/D(FP). If the test measurement standard deviation forMP denoted as SMP, is constant over this a to b range, theabsolute test sensitivity designated as cAis defined by Eq 1.cA5 | K |/SMP(1)The equation indicates that f
48、or the selected region of interest,test sensitivity will increase with the increase of the numerical(absolute) value of the slope,|K|andsensitivity will increasethe more precise the MP measurement. Thus, cAmay be usedas a criterion of technical merit to select one of a number of testmethods to measu
49、re the FP provided that a functional relation-ship, MP = f (FP), can be established for each test method.7.3.2 Absolute test sensitivity may not be uniform orconstant across a broad range of MPor FPvalues. It is constantacross a specified range, only if the direct (not transformed)MP versus FP relationship is linear and the test error SMPisconstant. With an assumed monotonic relationship between FPand MP, absolute test sensitivity, cA, may be evaluated on thebasis of (1) two or more CMs, (or objects) with differentknown FP values or (2) a theoretical relationship bet
