7HFKQLFDO 3DSHU - Troemner

1 Technical PaperINTERCOMPARISON BETWE E N CENAM, BIPM, AND Troemner TODETERMINE THE VOLUME magnetic susceptibility OF A 100 g WEIGHT1 INTERCOMPARISON BETWEEN CENAM1, BIPM2 AND TROEMNER3 TODETERMINE THE VOLUME magnetic susceptibility OF A 100 gWEIGHTJ Nava1, R S Davis2 and J Moran31 Centro Nacional de Metrolog a, CENAM, Bureau International des Poids et Mesures, BIPM, S vres Cedex, France3 Accredited Laboratory by NVLAP, magnetic interactions may lead to errors in mass measurements of high accuracy. Requirements ofmagnetic susceptibility and permanent magnetization on the mass standards (weights) have been included inthe new draft of OIML R111.

2 measurement methods and instruments for characterizing these magneticproperties of the weights have been developed as well. The BIPM susceptometer has been adopted by anumber of primary laboratories with responsibility for mass standards. We report a comparison of susceptibilityresults by three different laboratories using this IntroductionThe BIPM Susceptomer has been adopted by anumber of primary laboratories that they have theresponsibility of the mass standards of highaccuracy.

3 The new draft OIML R111 has includednew requirements on volume magnetic susceptibilityand permanent magnetization [1] for massstandards (weights). Volume magnetic susceptibilitycan be measured quantitatively using the BIPM susceptometer and permanent magnetization can , the weighing of a body using a weighinginstrument is based on the gravitational force on thebody. However, the presence of the magnetic forcesdue to magnetic properties of the body and alsofrom the ambient magnetic field generated by theweighing instrument (electrical circuits, coils andpermanent magnets) should not be considerednegligible [3].

4 The aim of this comparison is to check the use of thesusceptometer described in [2] to determine thevolume magnetic susceptibility of a Class E2 100 test procedure that CENAM, BIPM andTROEMNER used is proposed in [2] and is similar tothat described in [1].2. WeightOne 100 g weight was used to this comparison, itwas manufactured by Troemner . The 100 gweight complies with the OIML R111 requirements[4].The weight dimensions were given directly byTROEMNER and were checked by the others Model for magnetic Errors in Mass MetrologyThe z-component of a force on a sample with amagnetic susceptibility and magnetization M in amagnetic field is obtained from.

5 ()() =dVHMzdVHHzFz**200 (1)Where is the volume magnetic susceptibility of thestandard, M is its permanent magnetization (definedas the magnetic moment per unit volume in zerofield), H is the local magnetic field strength and thez-axis is parallel to the gravitational acceleration, g[2]. The integrals are taken over the volume of BIPM susceptometrerCENAM, BIPM and Troemner used thesusceptometrer described in [2], the difference wasthe weighing instrument adopted at thesusceptometrer, such characteristics are describe inthe following table 1:Table 1.

6 Weighing instrument characteristics adopted atthe susceptometer:InstituteWeighing instrumentCENAMMX 5 d= 0,001 mgBIPMUMT 5 d= 0,000 1 mgTROEMNERUMT 5 d= 0,000 1 mg2 Figure 1 shows the apparatus schematically, thedesign details also are described in [2]. View of the apparatus. A small rare-earth magnet is placedon a supporting column, which is on the pan balance. Anonmagnetic bridge, the height of which may be increased usingblocks, spans the weighing chamber. The body (weight) is placedon the TraceabilityCENAM determined the distance between themagnet and the bridge (Z0) [2] by using two differentmethods; one method was using a referencestandard of magnetic susceptibility whose value wasdetermined by the BIPM.

7 And the second methodwas using a set of gauge blocks and LVDT probehead these are traceable to the National LengthLaboratory in Mexico. Balance readings aretraceable to National Mass Laboratory (CENAM).The magnet was calibrated at the BIPM. Thedimensions (for the geometry correction) of theweights were given by determined the distance Z0 [2] by using areference standard of magnetic susceptibility whosevalue was determined by the BIPM with respect tocalibrated susceptibility standards obtained from anumber of national measurement institutes.

8 Balancereadings are traceable to Mass Section of the magnet was calibrated at the BIPM [2,6]. Thedimensions (for the geometry correction) of theweights were given by determined the distance Z0 [2] by usinga reference standard of magnetic susceptibilitywhose value was determined by the BIPM. Balancereadings are traceable to National Mass Laboratory(NIST). The magnet was calibrated at the BIPM. Thedimensions (for the geometry correction) of theweights were given by measurements were carried out as follows:The measurements were carried out byTROEMNER at March 2001, CENAM in April 2001and the last measurements the BIPM in May ResultsThe results of the comparison are presented in thefollowing table 2:Table 2.

9 Results of the comparisonTROEMNERM agnetic susceptibility 0,003 0U 0,000 6 magnetic flux density (max.)0,8 mTZ027,71 mmCENAMM agnetic susceptibility 0,003 1U 0,000 6 magnetic flux density (max.)1,4 mTZ022,16 mmBIPMM agnetic susceptibility 0,003 1U 0,000 3 magnetic flux density (max.)0,9 mTZ026,64 UncertaintyThe propagation of uncertainty from various sourcesis given in the following table 3, where ur(x)represents the relative standard uncertainty of aparameter x:Table 3.

10 Relative standard uncertainties of a parameter xSource ofUncertaintyContribution to ur( )ur(ma)ur(ma)ur(ma(s))()()smuYarf +21ur(m)()muYrfur( s)()srfuY +21 Where ma(s)*g = Fa(s), g is the local acceleration ofgravity and Fa(s) is the force between the magnetand the standard. Thus ma(s) is the average balancereading found when measuring the standard and mais the corresponding value of the unknown, m is themagnetic moment of the magnet, and s is themagnetic susceptibility of the standard Yf, which isapproximately 0,5 for these measurements, takesaccount of correlations between Z0, m, and s[6].