IMPROVEMENTS RELATING TO THE CALIBRATION OF THERMOMETERS
Abstract
ITS-90 specifies a series of temperatures which are used to define the scale. Supplementary Information
to ITS-90 describes how these substances can be embodied into structures, known as fixed point cells
that are useful to thermometrists.
At present, fixed point cells are used with apparatus such as cryostats, baths and furnaces. However,
current furnaces for example are expensive. The ideal furnace employs a heat siphon which has thermal
conductance 200 times better than copper to create isothermal conditions during phase transitions. It may
not be possible for a furnace to use the working fluid of choice for safety reasons. There is often poor
thermal conductance between the furnace and the fixed point cell such that a series of furnaces are
required to enable the required temperature range to be achieved.
A patent application filed during September 2007 describes a combined metal clad fixed point cell and
heat siphon, which when heated provides an isothermal environment for the metal within to change state.
The outer wall of the cell becomes the inner wall of the heat siphon with cost as well as performance
benefits. The device is called a “Siphonic Cell”. Siphonic Cells can be of Indium, Tin, Zinc, Aluminium,
Silver, Gold or Copper.
A fixed point cell is not long enough to eliminate heat conductance along the standard platinum resistance
thermometer calibrated in it. Currently, using long furnaces, heat shunts and reflective baffles an attempt
is made to reduce these losses.
A second patent describes a device to compensate for the stem conduction problems caused when a
thermometer under test is not sufficiently immersed into a fixed point cell. The device is called an
“Immersion Compensator”.
Its advantage is that it is independent of the cell/heat siphon and so can be set to give perfect
compensation for stem conduction. A combination of the two devices should offer improvements in
thermometer calibration.
This paper describes the successful development of the above ideas into a working prototype and the
results obtained during 2008 and 2009.
Introduction
The most accurate method of calibrating thermometers is in a series of fixed point cells as described in
ITS-90.
To get the best results from the melt or freeze of a fixed point cell it needs to be placed in an apparatus
which is gradientless. Unfortunately no fixed point cell is long enough to eliminate the stem conduction of
the thermometer and so the apparatus and cell are designed with heat reflectors, thermal shunts and
layers of insulation to reduce stem conduction in the thermometer during calibration.
The result is a complicated and expensive compromise, without proper scientific basis. During 2007 an
attempt was made to separate the components and recombine them into a more logical and scientific
solution.
Isothermal Technology Limited is registered in England. Company number: 1530620
Registered address: Pine Grove, Southport, PR9 9AG, England.
Directors: J.P. Tavener C.Eng., MSc., M.I.E.E; I.F. Tavener
E-mail info@isotech.co.uk
Pine Grove • Southport • PR9 9AG • England
Tel +44 (0) 1704 543830 • Fax +44 (0) 1704 544799
• Internet www.isotech.co.uk
Combining Cell & Apparatus
The ideal apparatus to surround a cell is a heat pipe or heat siphon. If the outer wall of a metal clad fixed
point cell also became the inner wall of the heat siphon then a very simple structure of ideal thermal profile
would result.
Firstly over 160 UKAS certificates evaluating quartz and metal clad cells were compared [1].
The analysis showed no detriment occurred to the metal in the cell provided the metal cladding was
properly prepared. A literature search also confirmed the above [2, 3, 4, 5, 6, 7, 8].
A heat siphon provider was approached who was prepared to share his manufacture process. Perhaps
not surprisingly the metals and metal preparation procedures were almost identical to those we use in
preparing metal cladding for fixed point cells.
It therefore seemed probable that such combinations would work.
The concept was patented and called a Siphonic Cell (S.C.)
Working Fluids
Water in the Heat Siphon was suitable for indium and tin, caesium or potassium for zinc and sodium for
aluminium.
Immersion compensation (I.C.)
The depth from metal surface to the bottom of the re-entrant tube is 180mm and this is inadequate for
most SPRT’s. The unit under test therefore needs to go through an isothermal zone above the cell set to
the cells transition temperature. Called an Immersion Compensator this drilled and heated thermal block
sits on top the Siphonic cell.
The total depth required can be calculated as follows;
An 8mm diameter thermometer with 35mm sensing length requires maximum immersion at around 420°C
[9].
Traceable Temperatures (2) (P 136) shows an immersion of 28 diameters (K=2) gives 0.4mK error. The
new design has a total immersion of 300mm offering an immersion error of less then 0.1mK – 420°C [10].
Losses
The Siphonic cell is dewar shaped and has no losses below the cell. After profiling the Immersion
Compensator there are no losses above the cell. Extremely long plateaus with extremely close settings of
the controller should be achievable.
Isothermal Technology Limited is registered in England. Company number: 1530620
Registered address: Pine Grove, Southport, PR9 9AG, England.
Directors: J.P. Tavener C.Eng., MSc., M.I.E.E; I.F. Tavener
E-mail info@isotech.co.uk
Pine Grove • Southport • PR9 9AG • England
Tel +44 (0) 1704 543830 • Fax +44 (0) 1704 544799
• Internet www.isotech.co.uk
Results
CCT/2000-13 “Optimal Realisations of the Defining Fixed Points of the ITS-90 that are used for Contact
Thermometry” was chosen as the standard to which the results would be compared, as this document
describes “…techniques that should be used when it is desired to achieve realisations at the highest levels
of accuracy and precision that can be expected with the best equipment presently available” [11].
The same document describes desirable features of the melt and freeze curves.
“The desired features of (a melt) are constancy of the temperature of the plateau over 75 – 80% of the
total (melt) curve to within 1mK.
…(during freezing) relative to the maximum of the freezing curve a depression of the fixed point
temperature at 50% of the sample frozen of a few tenths of a milliKelvin.
…a 99.9999% pure sample might have depressions as follows; 0.5mK for In, 0.3mK for Sn, 0.5mK for Zn,
0.7mK for Al and 1.1mK for Ag.
A duration of the plateau of at least 10 hours.”
“…the samples liquidus-point temperature obtained from a slow freeze and from a melt obtained following
a fast freeze should agree within 0.2mK.”
The cell should be sufficiently immersed to exhibit no stem conduction effects over at least 3cm from the
bottom of the re-entrant tube.
The Siphonic cells with their Immersion Compensator were assembled into simple desktop apparatus
400mm high with built-in controllers, see figures 1 and 2. The assemblage is called an ITS-90 Isothermal
Tower.
ITS-90 isothermal towers of In, Sn, Zn and Al have been evaluated. Eight to ten melts and freezes were
made with various settings of the Siphonic cell and also the Immersion Compensator.
Plateau lengths could be varied by offsetting the Siphonic Cell’s temperature from 30+ hours (offset 0.1°C)
to 4 hours (offset 0.7°C).
In an additional test insulation was added above the immersion compensator. This made no measurable
difference to the thermometers temperature but it did reduce the losses and hence power needed by the
Immersion Compensator.
Total power consumption is less than 400 Watts.
Coincidence 0.1 to 0.2mK
Length of plateau 10H
80% melt 1mK
50% freeze 0.5mK
CCT/2000-13
Indium
+
In Siphonic
Cell
9
9
9
9
Purity 99.9999% 99.9999%
Conclusion: the Indium Siphonic Cell complies to the requirements of CCT/2000-13 for Indium.
Isothermal Technology Limited is registered in England. Company number: 1530620
Registered address: Pine Grove, Southport, PR9 9AG, England.
Directors: J.P. Tavener C.Eng., MSc., M.I.E.E; I.F. Tavener
E-mail info@isotech.co.uk
Pine Grove • Southport • PR9 9AG • England
Tel +44 (0) 1704 543830 • Fax +44 (0) 1704 544799
• Internet www.isotech.co.uk
Loading...
+ 7 hidden pages
You need points to download manuals.
1 point = 1 manual.
You can buy points or you can get point for every manual you upload.