IEC 60247 Insulating liquids - Measurement of relative permittivity, dielectric dissipation factor (tan δ) and d.c. resistivity
5 Apparatus
5.1 Test cell
The same test cell can be used for measurements of permittivity, tan δ and d.c. resistivity. A cell suitable for these purposes shall meet the following general requirements.
5.1.1 The design of the cell shall be such as to allow easy dismantling for cleaning of all its parts and reassembling without significantly changing the capacitance of the empty cell. The design shall also permit the use of the cell at the required constant temperature and shall provide means to measure and control the temperature of the liquid to the required accuracy. This may be achieved by means of an externally heated oven/bath or alternatively by internal electrical heating of the cell.

5.1.2 The materials used in constructing the cell shall be non-porous and capable of withstanding the required temperature satisfactorily. The alignment of the electrodes shall not be influenced by change of temperature.

5.1.3 The surfaces of the electrodes in contact with the liquid under test shall have a smooth finish to make their cleaning easier. There should be no chemical interactions between the liquids under test and the electrodes. These should also be unaffected by the cleaning materials. Cells made of stainless steel have been found satisfactory for testing all types of insulating liquids. Aluminium or its alloys should not be used because they may be attacked by alkaline detergents.

NOTE Generally, plated surfaces have been found less satisfactory than solid metal electrodes. However, surfaces plated with gold, nickel or rhodium have been found satisfactory provided they are well plated and remain undamaged. Rhodium-plated invar has been found satisfactory and has the additional advantage of low thermal expansion. Nickel or gold-plated brass and nickel-plated stainless steel have also been used.

5.1.4 Solid insulating materials used to support the electrodes shall have a low tan δ and a high resistivity. They should not absorb, or be adversely affected by, reference liquids, test liquids or cleaning materials.

NOTE Fused silica is generally considered to be a suitable material for use as an insulating material in the cell. As a consequence of the difference between the coefficients of linear expansion of usual metals and of fused silica, a sufficient radial clearance is necessary between the joints. This clearance may decrease the accuracy of electrode gap spacing.

5.1.5 The distance across the surface of the liquid and across the solid insulating material, between the guard electrode and the measuring electrode, shall be great enough to withstand the test voltage used.

5.1.6 Any cell which meets the requirements given in 5.1.1 to 5.1.5 may be used. Examples of cells (not exclusive or exhaustive) which may be used with low viscosity liquids and up to 2000 V are shown diagrammatically in Figures 1 to 5.

In the three-terminal cells, provision is made for an efficient electrical guard system which adequately shields the measuring electrode. Three-terminal cells shall be preferred when the most accurate determinations of permittivity are to be made. If necessary, in the case of measurements for which special screening is required, a removable screening cup can be added and electrically bonded to the appropriate conductor of the coaxial cable used for the connection to the bridge (see Figure 2).

With two-terminal cells, the shield on the lead usually connected to the guard electrode shall be securely clipped to the cable insulation in order to prevent contact with any other surface. When these cells are used for resistivity measurements it should be established that the resistance across the insulators, with the cell empty, is at least 1 00 times the resistance which is measured through the liquid. For a.c. measurements corresponding values for tan δ shall be established.

For highly insulating liquids, the additional losses due to the insulators that support the electrodes may alter the measurement. For this reason, it is recommended to use test cells in which there is no bridge made from any solid insulating material between the two measurement electrodes. The dissipation factor of such an empty cell should be lower than 10(-6) at 50 Hz.

To minimize the effect of contaminants from the surfaces in contact with the liquid, it is recommended to use cells with a low ratio of electrode surface area to liquid volume (for example, <5 cm(-1).

5.2 Test equipment
Test equipment shall be adequate to maintain the test cell with +/-1 °C of the prescribed temperature and shall provide screened electrical connections to the cell. The test cell shall be properly insulated from the earthed enclosure of the test equipment.

5.3 Glassware
Borosilicate laboratory glassware such as beakers, graduated cylinders, pipettes, etc., shall be used. All glassware used in handling samples shall be cleaned at least to the standard specified in Clause 6, and carefully dried.

5.4 Measuring instrument for permittivity and tan δ
Any a.c. capacitance and tan δ measuring device capable of a measuring precision and resolution appropriate to the samples to be tested can be used.

Examples of a.c. capacitance bridges and testing circuits may be found in IEC 60250.

5.5 Measuring instrument for d.c. resistivity
Suitable instruments and test circuits are described in IEC 60093. Any apparatus with the precision and resolution appropriate to the samples to be tested may be used.

5.6 Time-measuring device
A time-measuring device, accurate to 0.5 s, shall be used for measuring the time of electrification.

5.7 Safety
DANGER - Ensure that equipment safety devices are functioning properly.