Module 4 : Breakdown in Liquid and Solid Dielectrics
Lecture 20 : Definition and Measurements of Intrinsic and Practical Breakdown strengths of Liquid dielectrics

Intrinsic Breakdown Strength

The intrinsic breakdown of a dielectric is defined as the highest value of breakdown strength obtained after eliminating all known secondary effects which may influence the breakdown. The concept of intrinsic breakdown is ideal. It is, however, extremely difficult to ascertain whether an observed breakdown was intrinsic or not. Intrinsic strength can be suitably interpreted as the field intensity at which the particles in the material are accelerated without any limit by the direct electrostatic action of the field.

The mean free path of an electron in insulating oils is very short, of the order of 10-6 cm. An extremely high field intensity is therefore required to begin the elementary process of ionization with free electrons having a very short life time. These elementary processes through which energy loss into the liquid takes place, determine the breakdown mechanism. Hence these processes decide the breakdown strength of a dielectric. The elementary processes of   ionization are as follows:

  • the most likely elementary process in case of hydrocarbon liquids is by excitation or molecular vibration, which is equivalent to thermal vibrations. The vibrational modes of a hydrocarbon are determined by elementary vibrations of the C-C and C-H molecular bonds. The frequencies of such vibrations lie in the infra-red region with corresponding energy quanta of about 0.2 to 0.4 eV (1 eV = 1.6 x 10 -19 J). These may lead to local growth of vapour phase of microbubbles.
  • the process of dissociation of molecules in neutral, low molecular, gaseous particles due to severe molecular vibrations which requires energy levels in the range 1.5 to 7 eV.
  • excitation of metastables which may lead to ionization in a few stages, requiring energy levels of the order of 1.5 to 10 eV.
  • ionization process involving scintillation of electrons accompanied with weak luminescence, indicating energy quanta of several eV; which is greater than 10 eV in some liquids.

The elementary processes mentioned above depend upon the molecular structure of a dielectric. These may occur spatially and timely together. However, there is no evidence of ionization by electron collision in liquids as in the case of gases. The electrode process may play its usual role as described.

For a mean free path of electron of the order of 10-6 cm and an energy of about 10 eV, the theoretically estimated field intensity required for ionization in insulating liquids is of the order of 10,000 kV/cm. On the contrary, a field intensity of only about 200 to 400 kV/cm may be sufficient to cause ionization by the elementary process of excitation of molecular vibration.

Electric strengths of some highly purified liquids and liquefied gases as given by Lewis in [3.5] are given in Table 20.1. These breakdown strengths appear to have been measured on very thin liquid films and represent nearest comparative values of intrinsic strengths.
Table   20.1  Electric Strength of highly purified liquids and liquefied gases, Lewis in [3.5]

Liquids
Breakdown Strength kV/cm
Good oil
1000-4000
Benzene
1100
Silicone
1000-1200
Hexane
1100-1300
Hydrogen
                                  >1000
Oxygen
                                  2400
Nitrogen
1600-1800
 
 
Practical Breakdown Strength of liquid dielectrics
The practical breakdown strength of insulating liquids is measured with help of standard electrodes having weakly nonuniform field between them.  The peak value of electrical breakdown strength of commercially available purified insulating oils is about 350 kV/cm. This is highly contradictory to the measured intrinsic strengths of the order of 1000 k V/cm and above. As mentioned above, it is because of the secondary effects which influence the breakdown strength. If the condition of the liquid is very bad (impure, contaminated, etc.) one may measure an electric strength as low as 10 kV/cm only.
In order to compare  the electric strengths of different insulating liquids or the same liquid in its different conditions, the methods of measurement of breakdown strength have been standardized by different standards. Simple methods are thus evolved to estimate the quantity of insulating liquids. The methods laid down by TGL-15077, IS-6792, VDE-0370 and IEC-156 are almost identical. According to these, an ac power frequency voltage is applied to 1 - 2.5 mm gap between  two identical electrodes, so called 'calottes', of dimensions shown in Fig 20.1. The IS recommends the electrodes to be made of brass having a good surface finish, whereas VDE recommends copper as electrode material. These electrodes are placed in a container of given size and filled with about 300 cc of the sample oil. The measurement is conducted at room temperature (20°C) by increasing the applied sinusoidal waveform alternating voltage at a rate of 2-3 kV/s upto breakdown. The rms value of the breakdown voltage Ub is measured. The Schwaiger factor of uniformity, η for the given gap and electrode configuration is 0.97, that is, the field can be considered to be almost uniform.

For each sample of insulating liquid, six measurements of breakdown voltages are recommended to be performed at an interval of 2 min. (VDE), 1-5 min. (IEC)  or 10 min. (TGL), stirring the liquid thoroughly after every breakdown. TGL recommends to consider arithmetic mean of the measurement numbers two to six, whereas VDE and IS recommend mean of all the six values to be taken into account. If the mean rms breakdown voltage is Ub in kV, the peak value of electric strength of the liquid is given as follows:
                                                                                              
 
Fig 20.1  Electrode design having η = 0.97 for the measurement of electric strength of commercial liquid dielectrics according to VDE-0370.
 

Besides the purity and clean conditions of the sample of oil, the breakdown strength strongly depends upon the moisture content and the temperature of the insulating oils.

Power frequency ac breakdown voltage/ field intensity of transformer oil samples having different moisture content were measured by Holle [3.2] with increasing temperature according to the standard  method described above, Fig.20.2

 
Fig 20.2   ac power frequency breakdown field intensity (rms) with increasing temperature of a transformer oil having different water contents in ppm, Holle [3.2]
 
These measurement curves illustrate the effect of moisture content in oil samples on their breakdown strengths. It is interesting to observe that for very low (<20 ppm) and also for very high ( >200 ppm) moisture contents the breakdown strength is more or less independent of the temperature. However, between 20 and 100 ppm level of moisture content, the breakdown strength of oil strongly depends upon its temperature.
 
1