E. A. Solomonik 19/06/03

36-WG11/Nberg/158

L. L. Vladimirsky, E. A. Solomonik HVDC Power Transmission Research Institute (NIIPT) COMPARISON OF THE SPECIFIC SURFACE CONDUCTIVITY AND ESDD APPROACHES TO ASSESSMENT OF SALT DEPOSITS ON NATURALLY POLLUTED INSULATORS 1. General The recent IEC 60815 draft submitted by the WG11 TC36 for revision comprises a number of statements which, in our opinion, should be further discussed. Presented below is a comparison of assessment of the natural polluting deposit’s characteristics using the specific surface conductivity ρ techniques and the equivalent salt deposit density (ESDD/NSDD) method. These two parameters, along with some others that are not considered here, are used widely as the standard pollution level (PL) criterion of areas for which external insulation must be selected. In Russia and ex-USSR, a wealth of ρ and, to a smaller degree, ESDD/NSDD measurement data for naturally and artificially polluted (NP and AP, respectively) insulators have been accumulated over four decades since the 1960s. These comments are based on generalized data reported in [1-6] and in some other NIIPT publications. Both ρ and ESDD/NSDD were measured during the following studies: - selection of insulation levels (creepage distances) on the basis of breakdown (flashover) characteristics and parameters of the pollution layer of NP insulators taken out of service; - determination of flashover voltages of NP insulators by indirect methods (from characteristics of the pollution layer); - flashover studies of AP insulators of various configurations; - flashover studies of non-uniformly AP insulators; - selection of a pollutant for AP tests of external insulation. By way of illustration, Figs. 1 and 2 give NIIPT’s findings of 1968-1970 insulation level studies for 132 to 500 kV transmission lines in saline soil areas of Egypt. Back in 1969, the specific surface conductivity ρ of the polluting deposit was substantiated to be the principal criterion of assessing the pollution level of AP and NP insulators by a USSR national conference on polluted insulation research. Since that time, methods of determining ρ and its values for various PL areas were standardized by many Russian mandatory documents dealing with tests, selection, and operation of external insulation and used for practical applications. The operational experience of insulation selected with the help of this criterion proved the decision to be totally correct. As for the ESDD/NSDD method, which is also recognized in Russia, it is considered but a supplementary technique of evaluating the artificial and natural pollution level and used for selection of insulation quite seldom. Long-term field and laboratory polluted insulation studies of various territories provided basis for a number of standards and methodological documents, still in force in Russia, related to measurements of pollution layer parameters of naturally polluted porcelain and glass insulators [7-11]. Similar Russian guidelines have been developed for composite (polymer) insulators as well, with, however, a limited bank of measured data. The standards specify the scope, times, and procedures of natural pollution measurements. The most accurate, although time-consuming, method of selection of insulators for natural pollution conditions is through use of data on their wet flashover (breakdown) voltages, which are most often in artificial pollution tests. Specific surface conductivity inputs make insulator selection less accurate while facilitating it considerably. Additional insulator selection criteria that are also used in Russia for natural pollution conditions are density of salt deposits and properties of precipitation-carried pollutants accumulated in special collectors.

2. Selection of insulation levels by breakdown voltage and specific surface conductivity of NP insulators Russian requirements on selection of outdoor insulation by breakdown voltages and specific surface conductivity of wet polluted insulators in natural and industrial pollution areas at low values of specific conductivity of atmospheric precipitation (ρv under 0.5 mS/cm) are shown in Table 1. Similar criteria are also available for areas with a high conductivity of atmospheric precipitation (ρv from 0.5 to 10 mS/cm). Table 1 Russian standards for determining the specific effective creepage distance of electrical insulation λe in natural and industrial pollution areas (ρv under 0.5 mS/cm), based on measurements of λ e - bd and

æ at NP λ e - bd , cm/kV over 1.2 to 1.5 over 1.5 to 1.9 over 1.9 to 2.3 over 2.3 to 2.7 over 2.7 to 3.3

ρ , µS over 1 to 3 over 3 to 6 over 6 to 12 over 12 to 20 over 20 to 30

Minimum λe, cm/kV for 110 kV and above 1.60 2.00 2.50 3.10 3.70

PL according to [11] 1 2 3 4 over 4

Designations used in Table 1 are as follows:

λ e - bd = specific effective breakdown creepage distance, cm/kV; λ e-bd =

L , where: U bd ⋅ K

L = creepage distance of the test object (insulator or insulating structure), cm; U bd = average breakdown voltage of the test object over a period of time when it remains steady (varying within 10%), kV; К = creepage distance efficiency factor for the test object according to [11]; ρ = test time-averaged specific surface conductivity of the polluting deposit, µS; ρ = К f /R s ,

where: Кf = form factor of the test object; R s = test-time averaged surface resistance of the moisturesaturated polluting deposit, MOhm. λe = specific effective creepage distance for outdoor insulators, cm/kV, as specified by Russian standards [12]; Those individual measured values of λe-bd that are more than 20% below the average λ e - bd are disregarded. Likewise ignored are those test values of ρ that differ from the average æ by more than 50% at ρ ≤ 15 µS and more than 30% at ρ >15 µS. Taken as the design value of λ e - bd for each field station is the larger of two values, viz. the testtime average or the maximum for abnormally adverse pollution or precipitation conditions, such as exhibited during large-scale blackouts. The same approach is taken in determining the design value of the specific surface conductivity ρ.

2

Owing to a fairly large scatter of the measured breakdown voltage and specific surface conductivity values, reliable data for each location can be obtained only from a sufficiently large number of measurements, which must be continue for at least two years, so as to cover the worst-case pollution conditions for the study area. With both λ e - bd and æ data available, the value of λe should be found from λ e - bd . The standard values quoted in Table 1 cover insulator strings and pin insulators of overhead transmission lines with steel and reinforced concrete supports or towers, as well as external insulation of electric apparatus and insulators of outdoor switchgear. The standards were based on and substantiated by generalized NP measurements and apparatus service records for areas with various environmental conditions. NP breakdown voltage values were used for plotting dozens of insulation level maps for a majority of ex-USSR areas featuring natural and/or industrial pollution. Such maps are generally divided into local and regional categories covering part or totality of the power system service area. Breakdown voltage values were also applied for selection of insulation levels of many planned or upgraded electric power facilities. In the latter case, the breakdown voltage approach to selection of insulators and determination of pollution level exhibits only one disadvantage of being highly labor-intensive: it involves time-consuming operations of mounting, removing, transporting and testing large numbers of insulators, to say nothing about the need for purpose-oriented test equipment, such as a powerful test transformer, an artificial wetting system, and some other items. Tests call for at least two years; in order to take off the insulators, the field pollution standard must be de-energized. The ρ approach to insulator selection is not as time-consuming and calls for a smaller number of test insulators. However, the ρ test duration is the same as in the breakdown voltage case. 3. Comparison of parameters of insulator polluting deposit

Numerous natural pollution studies, which were carried out by NIIPT in compliance with respective mandatory procedures, made it possible to correlate the breakdown voltage of insulators to both ρ and ESDD/NSDD for specific environment conditions. By way of illustration, quoted below are conclusive experimental findings for two test cases. Here breakdown voltages are expressed via the L specific breakdown effective creepage distance λ e - bd (see above, λ e-bd = ). U bd ⋅ K Case 1 deals with a dry tropical climate, saline soil area featuring abundant dew condensation. For several years, both breakdown voltage and polluting deposit measurements were recorded simultaneously on a large number of regular cap-and-pin porcelain insulators. The range of ρ, ESDD, and λ e - bd was found to be 2-50 µS, 0.01-0.2 mg/cm, and 1.25-3.0 cm/kV, respectively. Case 2, dealing with a humid tropical climate, sea coast area, involved long-term studies of regular cap-and-pin porcelain and glass insulators. The averaged findings are summarized in Table 2. These and many other findings by NIIPT permit the following conclusions: - averaged breakdown voltage values for a specific area or areas with similar environmental conditions can be deduced from both ρ and ESDD; - the scatter of measured values of both ρ and ESDD remains within the characteristics of a single PL; - even with a considerable scatter of experimental findings, the breakdown voltage/ρ relationships for different types and intensities of natural pollution make possible establishment of reliable regularities for insulators of an identical configuration; however, similar data on ESDD are unavailable.

3

Table 2 Averaged field-study findings for cap-and-pin insulators exposed to saline sea pollution

æ ,

λ e - bd , cm/kV

SDD, mg/cm2

ESDD, mg/cm2

over 0.025 to 0.05

over 0.025 to 0.04

over 0.08 to 0.18

1

over 0.05 to 0.10

over 0.04 to 0.06

over 0.18 to 0.30

2

over 0.10 to 0.18

over 0.06 to 0.10

over 0.30 to 0.60

3

µS

over 1.1 to over 3 to 6 1.3 over 1.3 to over 6 to 10 1.6 over 1.6 to over 10 to 1.9 18

Overall density of pollutant, mg/cm2

PL

Besides, it is shown by available data that use of the insulator-type-averaged polluting deposit parameters yields ρ/ESDD ratios that range widely depending on pollution type and level. By way of illustration, Table 3 shows such averaged ρ/ESDD (µS⋅cm2/mg) ratios for regular porcelain cap-andpin insulators versus PL and some pollution varieties. Table 3 Ratios of polluting deposit parameters of cap-and-pin insulators for some natural pollution varieties PL 1 2 3 4

Pollution type Sea salt

ρ/ESDD 145 160 175 –

Pollution type Soil salt

ρ/ESDD – 600 500 250

Pollution type Fly ash of coalburning thermal power plants

ρ/ESDD 50 – – –

Results of similar measurements in different operating environments show that the ρ/ESDD ratio depends heavily on the natural pollution variety and level, the insulator type, non-uniformity of pollution, climatic conditions of the study area, as well as some other factors. Thus selection of insulation levels with the help of such general-purpose parameters as ρ or ESDD can very often lead to very different results. It is shown by NIIPT’s experience that the ρ-based insulation selection procedure should be further improved through developing recommendations for reference insulators, separately at least for marine, soil, and industrial pollution.

4. Comparison of insulation selection approaches Summarized in Table 4 are pros and contras for using the ρ and ESDD/NSDD polluting deposit characteristics in selection of outdoor insulation. The data available at NIIPT permit the following conclusions: - use of polluting deposit parameters for selection of insulation levels dictates application of simple insulator shapes, as suggested by IEC WG11 TC36 in revision of IEC standard 60815; - selection of insulation levels by a common criterion for different pollution varieties, which is based on polluting deposit parameters, whether ЕSDD, as in the IEС 60815 under revision, or ρ, as in many other cases, can result in serious errors; - available Russian NP research findings generally give preference to the ρ-based approach to selection of insulation levels; use of the ESDD/NSDD approach calls for further data acquisition;

4

Table 4 Comparison of insulation selection approaches Parameter ρ Proved to be good with a majority of pollutants Comparatively easy to determine

ESDD/NSDD Proved to be good with marine and soil salt pollutants Fairly hard to determine

Quite often the test insulator must not be taken out of service Used widely in a number of countries

As a rule, the test insulator must be taken out of service Used widely in a number of countries

Acceptable for any insulator materials Applicable for all kinds of pollution and wetting conditions More accurate when used to determine needed insulation level by a uniform standard for areas with different pollutant deposits

Unacceptable for composite insulators Non-applicable for chemical pollutants and industrial/natural conducting wetting media Less accurate when used to determine needed insulation level by a uniform standard for areas with different pollutant deposits

- under a specific set of pollution conditions, some insulator configurations can be proved superior by the ESDD/NSDD criterion and others, by the ρ method; - insulation levels can be selected on the basis of both ρ and ESDD/NSDD data with appropriate experimental data available for an area with similar pollution and wetting conditions; - at the present stage of development of an international guide on selection of outdoor insulation in polluted areas, it is believed necessary to keep using both polluting deposit parameters because PL is traditionally determined in many countries, including Russia, with the help of ρ, while in others preference is given to ESDD; - it is thought practical to generalize and analyze, under auspices of IEC WG11 TC36, all available data obtained by various researchers in order to derive reference PL vs. ρ and PL vs. ESDD/NSDD relationships for a wide range of marine, soil, industry and other pollution varieties; it is suggested also that further NP research be carried out for identification of most appropriate application areas for each polluting deposit parameter. However, some suggestions on the issue under discussion are deemed possible even now.

5. Preliminary recommendations on application of ρ and ESDD/NSDD criteria under different conditions Because it is unlikely that breakdown strength test data for field-polluted insulators would be used on a large scale in selection of external insulation, the approach followed in revision of IEC 60815 and suggesting that PL be determined from parameters of the natural polluting deposit on the insulator surface is highly commendable. Table 5 summarizes NIIPT’s vision of using polluting deposit parameters for selection of insulation levels. With respect to composite insulators, the IEC 60815 standard should specify, regardless of pollution sources, the breakdown voltage of wet polluted units as the principal insulation selection parameter, to be supplemented by the criterion of specific surface conductivity of the polluting deposit on the units.

5

Table 5 Tentative suggestions for IEC 60815 on application of ρ and ESDD/NSDD for porcelain and glass insulators under different conditions Pollution type Saline soil Rural and non-industrial urban Sea spray without conducting wetting Sea spray with conducting wetting Solid industrial cementing pollution, pollution from coal/shale-burning thermal power plants Pollution from oil-burning thermal power plants Chemical industry pollution Petrochemical industry pollution

Principal IEC 60815 parameter ESDD, ρ ρ ESDD, ρ ρ

Additional IEC 60815 parameter NSDD ESDD/NSDD NSDD ESDD/NSDD

ESDD, ρ

NSDD

ρ

–

ρ ρ

– ESDD/NSDD

Recommendations on determining PL from ρ are given in Table 1. As for the PL versus ESDD/NSDD relationship, the recommendations of the draft IEC WG11 TC36 document appear to be acceptable, with the exception of cases shown in Table 5.

References 1. Merkhalev S. D., Solomonik E. A., Line and Substation Insulation in Polluted Atmosphere Areas. Leningrad, Energiya Publishers, 1973, 158 p. (in Russian) 2. Merkhalev S. D.. Solomonik E. A., Selection and Maintenance of Insulation in Polluted Atmosphere Areas. Energoatomizdat Publishers, Leningrad, 1983, 119 с. (in Russian) 3. Vladimirsky L. L.,. Chernevich L. V., Yakovleva T. V., Polluted line insulation studies on routes of planned long-distance power transmission lines. NIIPT Proceedings “Insulation of Overhead Lines and Switchgear in Polluted Atmosphere Areas”. 1983, pp. 15-20 (in Russian). 4. Merkhalev S. D., Vladimirsky L. L., Insulation of HVDC Overhead Lines. In “HVDC Power Transmission”, Energoatomizdat Publishers, Moscow, 1985, pp. 38-50 (in Russian). 5. Vladimirsky L. L., Selection of external insulation on the basis of field studies of soil pollution areas. NIIPT Proceedings “Insulation of 110 to 1150 kV Overhead Power Transmission Lines”, Leningrad, 1989, pp. 61-66 (in Russian). 6. Gurgenidze A. M., Solomonik E. A., Studies of polluting deposit parameters and breakdown characteristics of insulators in industrial areas. NIIPT Proceedings “Insulation of 110 to 1150 kV Overhead Power Transmission Lines”, Leningrad, 1989, pp. 56-60 (in Russian). 7. Instructions on Determination of Breakdown Characteristics of Naturally Polluted Insulators. Soyuztekhenergo, Moscow, 1977, 29 p. (in Russian). 8. Instructions on Determination of Characteristics of the Natural Pollution Layer of Insulators. Soyuztekhenergo, Moscow, 1978. 43 p. (in Russian) 9. Instructions on Plotting of Overhead Line and Switchgear Insulation Level Maps of Polluted Atmosphere Areas. Soyuztekhenergo, Moscow, 1985. 60 p. (in Russian). 10. National Standard GOST 10390-86. Electric Equipment rated at 3 kV and above. Test Procedures for Polluted External Insulation. Moscow, 1986, 23 p. (in Russian). 11. Regulations on Electric Installations. 7th edition. Chapter 1.9 “Insulation of Electric Installations”. Moscow, 2002 (in Russian). 12. L. L. Vladimirsky. E. A. Solomonik. Russian Guidelines for Selection of Outdoor Insulation. 36 to WG11/Cavtat/147 6

Fig. 2. String height-related flashover voltage of a string of PFE-11 insulators versus ESDD of NP deposit. 1,6 Eh, kV/cm 1,4

1,2

1

0,8

0,6

0,4

0,2 2

ESDD, mg/sm

0 0

0,02

0,04

0,06

0,08

0,1

0,12

0,14

0,16

0,18

0,2

Fig. 1. String height-related flashover voltage of a string of PFE-11 insulators versus specific surface conductivity of NP deposit. 2 Eh, kV/cm 1,8

1,6

1,4

1,2

1

0,8

0,6

0,4

0,2 X, µSm 0 0

10

20

30

40

50

60

8