DIN 51562-4-1999 Viscometry - Measurement of kinematic viscosity by means of the Ubbelohde viscometer - Part 4 Viscometer calibration and determination of the uncertainty of measur.pdf

1、DEUTSCHE NORM Januarv 1999 Viscometry Determination of kinematic viscosity using the Ubbelohde viscometer Part 4: Calibration of viscometers and determination of uncertainty of measurement DIN 51 562 -4 ICs 17.060 Descriptors: Viscometry, calibration, Ubbelohde viscometer, uncertainty of measurement

2、. Viskosimetrie - Messung der kinematischen Viskositt mit dem Ubbelohde-Viskosimeter - Teil 4: Viskosimeterkalibrierung und Ermittlung der Meunsicherheit In keeping with current practice in standards published by the International Organization for Standardization (ISO), a comma has been used through

3、out as the decimal marker. Foreword This standard has been prepared by Technical Committee Viskosimetrie of the Normenausschu Material- prfung (Materials Testing Standards Committee). This standard should be used together with DIN 51562-1 and (when using micro-Ubbelohde viscometers) DIN 51 562-2. Th

4、e apparatus requirements and measurement procedures covered in these two standards are not repeated here. The DIN 51562 series of standards comprises the following: Part 1 Part 2 Micro-Ubbelohde viscometers Part 3 Part 4 Apparatus and measurement procedure Relative change in viscosity at short flow

5、times Calibration of viscometers and determination of uncertainty of measurement 1 Scope This standard specifies the calibration of Ubbelohde viscometers as in DIN 51562-1 anc DIP 51562-2, as well as the determination of measurement uncertainty; these activities help increase the comparability of me

6、asurement results. When calibrating with standard viscometers or standard viscosity samples that are traceable to national standards, the requirements of DIN EN IS0 9001 are also met. 2 Normative references This standard incorporates, by dated or undated reference, provisions from other publications

7、. These normative references are cited at the appropriate places in the text, and the titles of the publications are listed below. For dated references, subsequent amendments to or revisions of any of these publications apply to this standard only when incorporated in it by amendment or revision. Fo

8、r undated references, the latest edition of the publication referred to applies. DIN 131 9-3 Basic concepts in metrology - Evaluating measurements of a single measurand and expres- sion of uncertainty DIN 51 562-1 Determination of kinematic viscosity using the Ubbelohde viscometer - Apparatus and me

9、asurement procedure DIN 51 562-2 Determination of kinematic viscosity using the Ubbelohde viscometer - Micro-Ubbelohde viscometers DIN 5301 2 Capillary viscometry of Newtonian liquids - Sources of error and corrections DIN 5301 7 Determination of temperature coefficient of viscosity of liquids Conti

10、nued on pages 2 to 8. Translation by DIN-Sprachendienst. In case of doubt, the German-language original should be consulted as the authoritative text. No pari of this translation may be reproduced without the prior permission of V Deutsches Institut fur Normung e. V., Berlin. Luth Verlag GmbH, D-107

11、72 Berlin, has the exclusive right of sale for German Standards (DIN-Normen). Ref. No. DIN 51562-4 : 1999-0 English price group 07 Sales No. 0107 11.99 Page 2 DIN 51 562-4 : 1999-01 DIN EN IS0 9001 Quality systems - Model for quality assurance in design/development, production, instal- lation and se

12、rvicing (IS0 9001 : 1994) IS0 3585 : 1991 Borosilicate glass 3.3 - Properties ISO/TR 3666 Viscosity of water 3 Concepts The concepts defined in DIN 51562-1 shall apply. 4 Symbols and units The symbols and units defined below in table 1 and in DIN 51562-1 shall apply. Table 1: Symbols, quantities and

13、 units 5 Apparatus and working substances 5.1 General The apparatus shall be as in DIN 51562-1; however, for calibration by direct comparison as described in subclause 6.4, the constant-temperature bath shall be large enough to accommodate at least three viscometers (two standard viscometers and one

14、 or more viscometers requiring calibration). 5.2 Standard samples and calibration liquids Calibration as in subclause 6.3 requires standard viscosity samples, each of which shall have a calibration certificate containing the following information: kinematic viscosity at the given temperature, uncert

15、ainty of viscosity values (expressed preferably in accordance with DIN 131 9-3, for a coverage factor, k, of 2, equivalent to a confidence level of about 95 YO), the temperature coefficient of kinematic viscosity at the given temperature (cf. DIN 5301 7), and the period of validity of the certificat

16、e. For calibration as in subclause 6.4, any Newtonian liquid with a suitable viscosity may be used, as long as it meets the following requirements. a) It shall be sufficiently transparent to allow visual assessment of the passage of the meniscus. b) There shall be no change in viscosity during testi

17、ng or calibration (e.g. due to evaporation of a compo- nent, water absorption, or heating carried out to facilitate filling of the viscometer). c) It shall be easily soluble in organic solvents that are non-toxic or of low toxicity which are used for cleaning the viscometer (e.g. using special boili

18、ng-point (SBP) spirit). Suitable Newtonian liquids are referred to in this standard as ?KF liquids?. These include pure organic liquids, mineral oils without additives, poly-a-olefines, and polyisobutylenes. Silicone oils are not suitable. Page 3 DIN 51 562-4 : 1999-01 O 6 Calibration 6.1 General Si

19、nce the glass used for producing Ubbelohde viscometers (borosilicate glass 3.3 to IS0 3585) has a very low coefficient of linear thermal expansion, the effects of temperature on the viscometer constant are negligible within the range of 1 O “C to 1 O0 “C (cf. DIN 5301 2). Therefore, calibration may

20、be performed at any temperature within this range, preferably, however, at room temperature (I 3 OC). Automatic measuring devices may be used; see subclause 8.5 of DIN 51562-1, January 1999 edition, for more detai Is. NP-1 600 to 800 NP-2 1000 to 1500 6.2 Cleaning of viscometers Viscometers shall be

21、 cleaned as described in DIN 51562-1. The viscometer to be calibrated should be cleaned using the procedure for heavily contaminated viscometers. Oc to Ia 6.3 Calibration with standard samples Determine the constants of viscometers of sizes II to V using two standard samples (denoted here as NP-1 an

22、d NP-2). For viscometers of sizes O to Ia, athird sample (denoted NP-O) shall be used to establish constant H of the kinetic energy correction (cf. DIN 51 562-1 regarding viscometer sizes). Standard samples shall have the flow times specified in table 2. Table 2: Flow times of standard samples for v

23、iscometers of different sizes NP-1 400 to 600 NP-2 800 to 1200 Viscometer I size no. as in I standard I Flow time. in s NP-1 NP-2 I DIN 51562-1 I sample I 300 to 500 700 to 1200 I NP-O I 350 to 450 I NP-O I 180 to 250 Carry out viscosity measurements as specified in DIN 51 562-1. For each sample, ca

24、lculate the mean flow time 4 t =1zti n (where n 2 5) and the relative difference between the longest and shortest flow times Et = tmax - tmin t whereby the following conditions must be fulfilled: 1 . 10-3 for Y I 1 O00 mm2/s 2 . 10-3 for Y 1 O00 mm2/s (3) Et If the conditions in (3) are not fulfille

25、d, clean the viscometer again and repeat the measurement using a new sample. If the conditions are still not fulfilled even after repeat measurements in accordance with DIN 51 562-1, the viscometer is not suitable for calibration as in this standard. Calculate the viscometer constant using the follo

26、wing equation: Y K=- t If condition (5) is met, IKl -KzI I 5 u, r7 (4) K2 then calculate Kas the mean of the values K, and K2, determined using standard samples NP-1 and NP-2, respectively; in this case, u; is the relative uncertainty of viscosities Y, and v2 (cf. subclause 7.2). If the uncer- taint

27、ies for the two values differ, the mean shall be taken. If condition (5) is not met, then clean the viscometers and repeat the measurements of samples NP-1 and NP-2. Page 4 DIN 51 562-4 : 1999-01 For viscometers of sizes O to Ia, the following conditions shall also be fulfilled: IKl -K2l lK1 -K2 1 2

28、 13. IKO -K2 1 IKo -K1l K2 K2 2 13. Kl K2 If condition (6) only, or conditions (6) and (7) are fulfilled, then calculate H using equation (8), rounding the result off to two significant figures. H = K2 .tO -yo .tg If H is positive, then use equation (4) to calculate the constant K, again, using the

29、corrected flow time: The viscometer constant K is then taken as the mean of K, and K2, to four significant figures. When using viscometers that have been calibrated as described above, viscosity is given by equation (1 O): Y=K t- KHt2 If H is negative, repeat the measurement of sample NP-O and, if n

30、ecessary, that of samples NP-1 and NP-2 as well. 6.4 Calibration by direct comparison with standard viscometers Standard viscometers used for calibration shall be of the same size number as the viscometers to be calibrated. Place two standard viscometers and the viscometer(s) to be calibrated in a c

31、onstant-temperature bath and fill them with the same calibration liquid (KF). Determine the viscometer constant using two calibration liquids (denoted here as KF-1 and KF-2). For viscometers of size numbers O to la, a third liquid (denoted KF-O) shall be used to establish the constant H. Calibration

32、 liquids shall have the flow times specified in table 2 for the corresponding standard samples. For each viscometer, measure the flow time for each calibration liquid using the method specified in DIN 51562-1, except that five measurements of flow times are to be made (.e. y1 = 5). Measurements shal

33、l be performed simultaneously in all viscometers, taking into consideration the time necessary to take the readings. The bath temperature need not be precisely adjusted to a given temperature, but the temperature is to be kept constant throughout the entire bath during the measurement. For each visc

34、ometer, calculate the mean flow time using (1) and the relative difference between the shortest and longest flow times using (2); this difference shall meet condition (3). If this is not the case, clean all the viscometers again and repeat the measurement with new calibration liquids. Calculate the

35、viscosity of calibration liquid KF-1 for each of the two standard viscometers (vN1 and vN2) using the method described in DIN 51 562-1, taking the measured flow time and, if necessary, the kinetic energy correction (as stated in the calibration certificate or other documentation). The relative diffe

36、rence, E, is to be calculated using the mean value Y of vN1 and vN2: lYN1 -YN2 1 Y E, = If the following condition is not fulfilled, repeat the measurement after cleaning and refilling the standard viscometers and those to be calibrated: E, I uK (1 2) where u; is the uncertainty of the viscometer co

37、nstant (cf. subclause 7.2). If the relative uncertainty differs for each constant, then the mean uncertainty shall be taken for u; in (1 2). Calculate the viscosity of calibration liquid KF-2 in the same manner. Using the viscosity values obtained with the standard viscometers, calculate the viscome

38、ter constants accord- ing to (4) and (1 O), as the mean of both measurements. For size O to Ia viscometers, include the kinetic energy correction as in subclause 6.3. 7 Uncertainty of measurement 7.1 General The viscosity of water at 20 “C serves as the internationally accepted basis for viscosity m

39、easurements (cf. ISO/TR 3666). However, the uncertainty of this basic value is not considered here in the expression of uncer- tainty. Page 5 DIN 51 562-4 : 1999-01 5 5 The method of determining uncertainty specified here is in accordance with DIN 1319-3 (cf. Appendix A). The uncertainties are given

40、 for a coverage factor of 2, which corresponds to a confidence level of approxi- mately 95 YO. I. 10-3 3,7.10-8 2.10-3 14,9 .IO- 7.2 Uncertainty of viscosity measurement The contributions of the following influence quantities shall be considered when calculating uncertainty. It is assumed in the fol

41、lowing that these quantities are the same for each viscometer; if this is not the case, then the mean variance shall be taken. a) Relative variance of viscometer constant, (s;) If the relative uncertainty of the viscometer constant, u;, is given fork = 2 (e.g. in the calibration certificate), then t

42、he variance is given by equation (13) (cf. Appendix A): 1 (s;02 = 4b;d2 (1 3) If this uncertainty is given only as a limiting value, then equation (1 4) applies: b) Relative variance of results obtained by timing device, (stimer)2 It is assumed that, as in common practice, two timers have been used

43、simultaneously, one for each viscom- eter to be calibrated. If these timers meet the requirements of subclause 8.3 of DIN 51 562-1, then the variance is given by: 21 2 ( siher) = - - (2 - 10-4) = 1,3.10-8 3 c) Relative variance of temperature measurement, The contribution of temperature measurement

44、to the uncertainty of the viscosity measurement includes the calibration uncertainty of the temperature measuring device at the measurement temperature, as well as the temperature gradients in the bath; this contribution is expressed by = (st? .U,) (1 6) where U, is the temperature coefficient of vi

45、scosity (cf. DIN 53017) of the liquid under test; for sg, see table 1. If the requirements in subclauses 8.2 and 8.4 of DIN 51 562-1 are met, then for U, I 0,l K-l 1 3 (s;,) = -(0,03.0,1)2 = 3-10-6 d) Relative variance of viscosity due to inclination error, SN) See DIN 53012 for determining the incl

46、ination error. If the requirements in subclauses 8.1 and 8.4 of DIN 51562-1 are met, then e) Relative variance of flow time, (sJ2 This quantity comprises the variances in the visual (or automatic) recording of the passage of the meniscus, as well as changes in temperature over time. These variances

47、shall be calculated as in Appendix A. If the requirements in clauses 12 and 13.2 of DIN 51 562-1 are met, then (sJ2 may be estimated on the basis of equation (A.7) and table 3. Table 3: Relative variance of flow time as a function of y1 and E+ For the relative variance of a viscosity value determine

48、d using two viscometers, each having its own ther- mometer (cf. Note in Appendix A), use (1 9): (1 9) 1 (si)2 = (s; +Te (s:irnerI2 + (sk,v)2 + (SN) + ($11 If the measurements obtained have different variances (e.g. when using viscometers with different uncer- tainties for the viscometer constants),

49、the mean variance shall be taken. Page 6 DIN 51 562-4 : 1999-01 For a coverage factor of 2, the relative uncertainty of measurement is u, = 2 . s, NOTE: For viscometers of sizes O to Ia, add 0,5. At,/t to u, (linear addition) in order to account for the uncertainty and systematic error of the kinetic energy correction, At, (cf. equation (9). 7.3 Uncertainty of viscometer constant determined using standard samples It is assumed that the relative uncertainty is the same for each standard sample (NP-1 and NP-2); otherwise, the mean shall be taken. Fo