| 公司介绍 | 产 品 | 实验室 | 技术图书馆 | 培 训 | 新动态 | 联系我们 |

 


Home

Company

Products

Lab

Tech Library

Training

What's New

THE WATER TRIPLE POINT
AND GALLIUM POINT
IN SECONDARY LABORATORIES
IN GERMANY

By Peter Klasmeier

ABSTRACT

Many secondary Laboratories in Germany, such as those that operate calibration stations of the German Calibration Service, or those that function as the measurement base for public utilities, are required to conform to regulations regarding their practice and procedures. These Laboratories are required to maintain the water triple point, and find it convenient to maintain the gallium point, in support of their calibration systems. However these are not used at frequent intervals, and can be realized in simpler and less costly equipment than in primary laboratories. This paper describes the use of the Isotech OCEANUS dry calibration bath with water triple point and gallium melt point cells.

DISCUSSION

In Germany, the National Laboratory, responsible for maintaining the International Temperature Scale, is the Physikalisch-Technische Bundesanstalt (PTB). Like many European National Laboratories, and unlike the United States practice, the PTB does not calibrate or recalibrate all thermometers in the hands of users, but, as a generality, calibrates only thermometers owned by the secondary laboratories which serve commercial and industrial users.

One such group of secondary laboratories are those accredited as calibration stations (Kalibrierstellen) of the Deutsche Kalibrierdienst (German Calibration Service; DKD). DKD Laboratories are accredited and supervised by the PTB, and are authorized to perform specific ranges of calibrations for users, and to issue legally-recognized calibration certificates. Usually the DKD Laboratories maintain appropriate fixed points of the ITS-90. (Many other countries have equivalent calibration services; for example, NAMAS, in England. Isothermal Technology is a NAMAS Laboratory).

Another such group of secondary laboratories are those that maintain the quality and accuracy of temperature measurement for public utilities. In many newer cities and towns of Germany, heat to buildings and homes is supplied from a central heat-generating facility, and distributed, usually as steam [1]. The distribution system is known as the Fernwrmedienst, or remote heating service. Accuracy of temperature measurement is required for the computation of heat use by the subscriber, who is charged on this basis. The range of calibrations performed by the Fernwrmedienst Laboratories is commonly limited to 0 to 140 deg C.

Both types of Laboratory are required to maintain the absolute thermometric standard represented by the water triple point, as a reference base for their working thermometers. The water triple point alone does not provide an assurance of calibration slope; two points at least are required for that; and with the introduction into common use of the melting point of gallium, a slope assurance can be obtained.

The water triple point is the equilibrium state of pure water in its liquid and solid phases under its own vapor pressure, and is 0.01 deg C, with an uncertainty of less than 0.000 150 K [2]. It is established in a familiar pattern of glass cell, which contains high-purity water sealed under its own vapor, and provides a re-entrant well for the thermometer under test. To prepare the cell for use, a portion of the water is frozen into the solid phase, and in traditional practice, this is done by introducing into the thermometer well a charge of liquid nitrogen or crushed dry ice. This procedure is tedious to carry out, and requires liquid nitrogen or dry ice as consumables. Therefore it is not ideally suited to Laboratories that prepare and use their cells only occasionally.

A method described by the National Physical Laboratory of England (the "slush method") is simpler for these Laboratories. Like all pure materials, pure water will exhibit undercool; that is, when cooled it will remain in its liquid phase at temperatures below that at which it would melt if it were in the solid phase. In the slush method, the water in the cell is cooled as a liquid to -6 deg C. When the cell is removed from the cooling environment and shaken, solid crystals form immediately throughout the liquid. If the cell is returned to the cooler and held at -6 deg C for another hour, a solid mantle of ice will form around the thermometer well. If the cell is then placed in an environment at 0.1 deg C, the triple point equilibrium can be maintained for several days. When the cell is to be used, the ice mantle is freed by inserting a warmed rod into the thermometer well, so that a thin film of liquid forms between the mantle of ice and the surface of the well.

The water triple point, as a calibration reference, is important to the secondary laboratory whether or not it is required by National law. First, it is the denominator of the ratio W in which thermometer calibrations are reported:

W = R(at temperature) / R(at water triple point)

and, as such, is necessary for the most precise use of the reference thermometer tables. Best practice is to measure the thermometer resistance at the water triple point immediately after each measurement of resistance at another temperature, and use these results in the W equation. Second, a continuing record of the resistance at the water triple point, maintained on a control chart, is an excellent quality control check on the stability of the reference thermometer.

The Isotech Oceanus dry-block bath is the ideal environment in which to perform the cooling and maintenance functions. A temperature range of -15 to +75 deg C is accomplished by Peltier solid-state modules, used for cooling and for heating. A microprocessor-based digital controller with a displayed resolution of 0.1 deg C, and with optional RS422 communications, establishes accurate and repeatable temperature levels.

The melting point of gallium provides a second absolute temperature reference at 29.7646 deg C, with an uncertainty of about 0.000 200K [3]. The gallium metal in the cell is first assured to be in the solid state, by immersing the bottom of the cell in crushed ice to a depth of 4 to 5 cm. This limitation of depth assures that the gallium will be frozen from the bottom of the cell upward; a necessary precaution, since gallium expands as it freezes. The frozen state can be demonstrated by measuring the temperature in the thermometer well, which, when the gallium is solid, will be in the vicinity of ambient temperature and well below 29 deg C.

The frozen cell is then placed in the Oceanus block at a controller setting of 30.1 deg C. As the cell warms it will reach the temperature where the gallium begins to melt, at which point the liquid-solid equilibrium has been achieved. In the Oceanus block, the melting equilibrium can be maintained for a number of hours.

Full realization of the ITS-90 to higher temperatures requires, of course, additional prescribed fixed points; the freezing points of indium, tin, zinc, aluminum, and for the full platinum resistance thermometer scale, silver. However, because of the extreme certainty of the water triple point and the gallium melting point, used in Oceanus, it is possible to use only these two points in the interpolation equation, and extrapolate to 660 deg C with an uncertainty of less than 10 mK [4]

The Oceanus dry well bath, combined with a water triple point cell or a gallium melt point cell, is the only apparatus necessary for the realization and maintenance, in the secondary laboratory, of two fundamental and defining fixed points of the ITS-90; absolute calibration standards. It is achieving wide use in this context, and has proven to be a versatile and economical apparatus.

Also, the Table of Footnote [4] below shows that, over much of the range of ITS-90, calibrations of thermometers within what may be industrially-acceptable limits of error can be obtained using only the water triple point and the gallium point.


Footnotes

[1] Remote heating of a number of buildings by steam from a central plant has been a common place in the commercial building areas of New York City for at least 50 years. I have also encountered this system in the newer cities of The Netherlands, that are built on land reclaimed from the Zuider Zee (HES).

[2] For a complete discussion of the water triple point, see Isotech Journal of Thermometry, Vol. 1 No. 1 pp 23-24 and Vol. 1 No. 2 pp 63-64.

[3] For a complete discussion of the gallium melting point, see Isotech Journal of Thermometry, Vol. 1 No. 1 p 24, and Vol. 1 No. 2 pp 59-60.

[4] In the Table which follows, Column 2 has been developed by calculating a table for an ideal thermometer, using the ratios which are the values of the reference function given on ITS-90 at the triple points of argon and mercury, the water triple point, and the freezing points of tin, zinc and aluminium, and employing the ITS-90 interpolation algorithm. Column 3 has been developed by using the ITS-90 algorithm given for the range 0 deg C to the melting point of gallium, and two fixed points only; the water triple point adjusted so that its resistance value is 120mK lower than nominal, and the gallium melting point adjusted so that its resistance value is 120mK higher than nominal, to represent readings in error by those amounts. Column 3 thus represents a very long extrapolation of the thermometer calibration in both directions, which includes the effects of these uncertainties of the two fixed points. Column 4 shows the differences between the extrapolated table of Column 3 and the ideal table of Column 2 in terms of ratio. Column 5 expresses these differences as Celsius degrees.



                                

  1       2       3            4          5

TEMP   ITS-90    WA-GA     WA-GA MINUS  WA-GA  MINUS

 C     RATIO     RATIO     ITS-90 RAT   ITS-90 DEG C



-200   .169754  .175975     .006221     .68665

-160   .342638  .354401     .002763     .12435

-120   .511547  .512280     .000733     .04926

- 80   .676790  .676901     .000111     .00969

- 40   .938436  .839431    -.000005    -.00050

   0   .999960  .999960     .000000     .00000

  40  1.158530 1.158535     .000005     .00005

  80  1.315171 1.315181     .000010     .00105

 120  1.469898 1.469913     .000015     .00168

 160  1.622724  1.622744    .000020     .00213

 200  1.733663 1.773688     .000025     .00268

 240  1.922729 1.922758     .000029     .00313

 280  2.069933 2.069967     .000034     .00368

 320  2.215285 2.215323     .000038     .00411

 360  2.358789 2.358832     .000040     .00463

 400  2.500441 2.500488     .000047     .00504

 440  2.640233 2.640285     .000052     .00551

 480  2.778149 2.778205     .000056     .00587

 520  2.914168 2.914229     .000061     .00632

 560  3.948268 3.048332     .000064     .00654

 600  3.180425 3.180493     .000068     .00690

 640  3.310618 3.310691     .000073     .00736

 680  3.438842 3.438909     .000077     .00755           







The ITS-90 calculations were done using the Isotech "Daedalus"

 MS-DOS program for interpolating ITS-90.



      

关于作者

Peter Klasmeier是德国Klasmeier Kalibrier - und Messtechnik GmbH的创始人兼董事长。他1973年在Darmstadt的The Technical Universityf获得了Dipl-Ing学位,之后在Fulda被Jumo雇佣,作为温度传感器工程师,负责设计、生产和市场开发。他在1986建立了他自己的公司。Peter一直活跃于几个DIN委员会,撰写温度标准。

© 2007 Isothermal Technology Ltd V1.005

Home Page Primary Lab Secondary Lab Industrial Calibration InfraRed Calibration