Insulation Coordination · External Insulation

Dielectric Strength of Overhead-Line External Insulation

External-insulation design for overhead transmission lines must weigh more than one kind of electrical stress. The required clearance or insulator arrangement may be set by switching overvoltages, by lightning overvoltages, by power-frequency voltage under contaminated conditions, or by the atmospheric corrections that apply to all of them. For many EHV and UHV lines switching controls the design; at lower voltages or in high-lightning areas, lightning performance dominates; in polluted or coastal environments, contamination under power frequency can be decisive. This guide explains how external-insulation strength is assessed for each stress, what parameters control it, and how the simplified CFO and gap-factor formulas should — and should not — be used. It is the umbrella over the APS notes on switching-surge strength and air-gap flashover modelling.

Reading time ≈ 34 min · External-insulation strength guide

The same physical clearance can have a different withstand under switching impulse, lightning impulse, power-frequency voltage and polluted-wet conditions — so external-insulation strength is not a single number. This page sets out how that strength is assessed across all the relevant stresses; the discharge physics and EMTP® flashover devices themselves are covered in the air-gap modelling note, and the switching case in detail on the switching-surge strength page.

Abbreviations used on this page

CFO / U₅₀Critical (50%) flashover voltage

BILBasic lightning impulse insulation level

BSLBasic switching impulse insulation level

\(k_g\)Gap factor (relative to rod-plane)

\(S\)Strike distance / gap spacing, m

\(\delta\)Relative air density

\(K_a\)Atmospheric correction factor

\(E_{50}\)Average CFO gradient, U₅₀/d, kV/m

\(\sigma_f\)Standard deviation of flashover voltage

CWFCritical wavefront

Key idea

External-insulation strength is not one number — switching, lightning and power-frequency (contamination) are separate design conditions, each with its own waveform, polarity, geometry and weather dependence.
Switching often controls EHV/UHV clearance (slow-front, positive polarity, near the critical wavefront); lightning is assessed via CFO/BIL and volt-time; clean power frequency rarely controls, but contamination can.
Use gap-factor and CFO formulas for screening, but strength is statistical (CFO + \(\sigma_f\)) and must be compared with the statistical overvoltage at the geometry-controlling strike distance.
Apply atmospheric / altitude correction where required and state its direction; BIL, BSL and CFO are related but distinct — standard withstand levels versus the 50% flashover value.

Key terms used on this page

01External insulation

Air clearances and insulator surfaces exposed to the atmosphere, whose strength depends on weather as well as geometry.

02CFO / U₅₀

The crest voltage giving a 50% probability of flashover for a defined waveform, geometry, polarity and weather.

03BIL

Basic lightning impulse insulation level — a specified standard lightning-impulse withstand level.

04BSL

Basic switching impulse insulation level — a specified standard switching-impulse withstand level.

05Gap factor \(k_g\)

The ratio of a configuration’s flashover voltage to that of the reference rod-plane gap (\(k_g=1\)).

06Volt-time characteristic

The relation between the voltage reached before breakdown and the time-to-breakdown; important for non-standard impulses.

07Non-standard impulse

A lightning overvoltage differing from the 1.2/50 µs test wave — short tails, multiple peaks, oscillations.

08Contamination flashover

Surface flashover of a polluted, wetted insulator at normal power frequency, without a transient overvoltage.

09Relative air density \(\delta\)

Air density relative to standard conditions; lower at altitude, reducing air-gap strength.

10Atmospheric correction \(K_a\)

A factor converting flashover/withstand voltage between standard and site atmospheric conditions.

11Creepage distance

The shortest path along the insulator surface between electrodes; key to contamination performance.

12CFO gradient \(E_{50}\)

CFO per unit gap length (kV/m); for positive rod-plane it is roughly constant at ~525 kV/m.

Section 1

What controls external-insulation design

External insulation must satisfy several stresses, and the controlling one varies with the line. For EHV/UHV lines, switching overvoltages usually set the clearance, because the slow-front strength saturates with spacing and can fall below the lightning strength. For lower-voltage lines or high-lightning areas, lightning performance governs the outage risk. For polluted or coastal sites, power-frequency contamination performance can dominate — flashovers occur in wet, dirty conditions with no transient at all. A complete assessment therefore considers switching-surge strength, lightning-impulse strength, power-frequency strength (clean and contaminated), phase-to-ground and phase-to-phase stress, tower and conductor geometry, insulator arrangement, wet/dry state, altitude and atmospheric correction, and the statistical nature of both overvoltage and strength. A single clearance value quoted without those conditions is incomplete.

Section 2

The inputs for an external-insulation assessment

The strength depends on three groups of parameters:

Geometrical — strike distance, conductor-to-crossarm and conductor-to-tower clearances, insulator-string length and type, tower-window geometry, phase position, V-/I-/strain arrangement, conductor and hardware shape.
Electrical — overvoltage type, waveform, time-to-crest, time-to-half, peak voltage, polarity, phase-to-ground and phase-to-phase stress, trapped charge, statistical overvoltage distribution.
Environmental — relative air density, humidity, rain, wetting, pollution, salt contamination, altitude, ice, snow and wind.

Section 3

Switching-surge strength (in brief)

Switching surges are slow-front overvoltages (line energisation, reclosing, trapped charge, reactive-plant switching). The standard switching impulse is 250/2500 µs, but real surges are produced by the network and should be modelled directly. Three features dominate the strength: the active front (the discharge develops in the part of the wave from about 70% of crest to crest, so peak voltage alone is not enough); the U-curve (for a fixed spacing the flashover voltage passes through a minimum at a critical wavefront, and a longer or shorter wave is not automatically more or less severe); and polarity (positive is the lower, controlling case for the non-uniform fields of line insulation). The dedicated switching-surge strength page develops these, plus geometry, phase position, weather and the worked design example; the essentials are summarised here so this umbrella page stands alone.

Section 4

Switching CFO and the statistical check

For preliminary positive-polarity, dry switching assessment, the CFO can be estimated from the gap factor \(k_g\) and strike distance \(S\). The Paris–Cortina form (and the IEC 60071-2 standard-impulse approximation — IEC 60071-2 being the insulation-coordination application guide) is \(\mathrm{CFO}=500\,k_g\,S^{0.6}\); for the critical wavefront, Gallet’s expression is used:

\[ \mathrm{CFO} = \frac{3400\,k_g}{1 + 8/S}\qquad\quad \mathrm{CFO} = 1080\,k_g\,\ln(0.46\,S + 1) \]

\(\mathrm{CFO}\): critical (50%) flashover voltage, kV crest
\(k_g\): gap factor (1 for rod-plane)
\(S\): controlling strike distance / air clearance, m — not necessarily the physical insulator string length \(S_I\) if a shorter conductor-to-structure path governs
\(\sigma_f\): standard deviation of flashover voltage, kV
\(\sigma\): per-unit coefficient of variation, \(\sigma_f/\mathrm{CFO}\) (\(\approx5\%\))
\(V_3\): statistical strength, \(\mathrm{CFO}(1-3\sigma)\), kV
\(E_2\): 2% statistical overvoltage, \(\mu_S+2.054\,\sigma_S\), kV
\(\mu_S\): mean switching overvoltage, kV
\(\sigma_S\): standard deviation of switching overvoltage, kV

Gallet (left) is for positive polarity, dry, to ~15 m; the \(k_g\) multiplies the 3400 and the \(8/S\) term uses the bare spacing — it is not \(3400/(1+8/(k_gS))\). The IEC rod-plane log form (right) applies to ~25 m. Strength is statistical: it is the 50%-flashover CFO with a standard deviation \(\sigma_f\) (coefficient of variation \(\sigma_f/\mathrm{CFO}\approx5\%\), broadly constant with distance). Design compares a statistical strength \(V_3=\mathrm{CFO}(1-3\sigma)\) with a statistical overvoltage \(E_2=\mu_S+2.054\,\sigma_S\) (the 2% value), requiring \(V_3\ge E_2\) — a simplified coordination check, not a literal flashover rate. The full treatment and a worked 400 kV example are on the switching-surge page. These gap-factor CFO expressions are empirical screening fits, valid within their stated distance and positive-polarity range — not a universal flashover model.

Section 5

Phase-to-ground and phase-to-phase stress

Switching surges stress both phase-to-ground and phase-to-phase insulation. Phase-to-ground compares each phase to ground or structure; phase-to-phase is the instantaneous difference of two phase voltages, and its strength depends on how the total is shared between positive and negative components:

\[ V_{AB}(t) = V_A(t) - V_B(t) \qquad\qquad \alpha = \frac{V_-}{V_{\text{tot}}} \]

\(V_{AB}\): instantaneous phase-to-phase voltage
\(V_A,\ V_B\): instantaneous phase-to-ground voltages
\(\alpha\): sharing factor \(V_-/V_{\text{tot}}\)

\(\alpha=0.5\) (equal-and-opposite components) is the reference condition for the phase-to-phase strength characteristic; the governing service case is where the actual stress — usually positive-dominated, \(\alpha<0.5\) — meets the strength curve, so \(\alpha=0.5\) is not automatically the worst case. For phase-to-phase studies, extract the actual maximum phase-to-phase voltage from the time-domain simulation; its worst case need not coincide with the highest phase-to-ground voltage on either phase.

Section 6

Lightning-impulse strength

Lightning strength is assessed differently. Lightning overvoltages are fast-front, and the actual voltage across the insulation may differ from the standard 1.2/50 µs laboratory impulse — travelling waves can be non-standard, multi-peaked, short-tailed or distorted by reflections, tower response, conductor coupling and arresters. For application, three pieces of data are normally needed: the CFO (or \(U_{50}\)), the standard deviation \(\sigma_f/\mathrm{CFO}\), and the volt-time characteristic. The lightning standard deviation is smaller than for switching — about 1% of CFO for positive air-gap impulses and ~3.6% for negative, rising to 5–9% where insulators strongly influence the CFO. Lightning strength is often characterised by a single value (CFO or BIL): voltages below it are treated as a 0% flashover probability and above it as 100%.

Section 7

Standard lightning-impulse CFO

For standard lightning impulses, simplified IEC-type expressions estimate the CFO of common configurations once the gap distance and gap factor are known:

\[ \mathrm{CFO}^{+} = 530\,S\,(0.74 + 0.26\,k_g) \qquad\qquad \mathrm{CFO}^{-} = 700\,S \]

\(\mathrm{CFO}^{+},\ \mathrm{CFO}^{-}\): positive / negative-polarity CFO, kV crest
\(S\): strike distance, m
\(k_g\): gap factor

The positive form applies up to about 10 m; the present IEC 60071-2 general negative formula is \(\mathrm{CFO}^{-}=700\,S\) (an older form is \(\mathrm{CFO}^{-}=950\,S^{0.8}(1.5-0.5\,k_g)\) for \(k_g\) in 1–1.44, and \(741\,S^{0.8}\) above). The CFO-versus-distance relation is essentially linear for positive polarity and the average positive rod-plane gradient \(E_{50}\) is roughly constant at ~525 kV/m, independent of clearance; for negative polarity it is non-linear and the gradient decreases as clearance grows. Positive rod-plane CFO is much lower than negative, so positive is usually controlling — though for gap factors above ~1.4 (uncommon on lines) the negative CFO can fall below the positive. Representative CFO gradients for practical configurations:

Table 1 — Representative lightning-impulse CFO gradients (kV/m) for practical line configurations, values at ~4 m (after Hileman, Table 2.17). The gradient is non-linear with distance.
Configuration	Positive (kV/m)	Negative (kV/m)
Outside arm — no insulators	600–625	600–625
Outside arm — with insulators	500–520	595–620
Conductor–upper structure	560–575	610–625
Conductor–upper rod	500–655	585–595

Section 8

Polarity and geometry under lightning

Geometry modifies the simple rod-plane picture. For a conductor-to-crossarm gap without insulators the CFO is essentially independent of polarity; with insulators in the gap, the positive CFO is very close to the negative, and the presence of insulators tends to reduce the negative-polarity breakdown voltage. Cap-and-pin strings are sensitive to the stress distribution along the string (grading/shielding rings reduce the stress on the end units); post, long-rod and composite insulators, with fewer metal parts, are less affected. For some arrangements (conductor-upper-structure, conductor-crossarm) the strength is close to that of the bare air gap. The practical consequence is that tower insulation strength should be assessed using the actual configuration, not a generic rod-plane formula, and that for large clearances testing is advised. As with air gaps, rain has only a secondary effect on lightning-impulse flashover.

Section 9

Volt-time characteristics

Lightning strength is often described by a volt-time curve relating the breakdown voltage \(V_B\) to the time-to-breakdown \(t\). Unlike the switching impulse (governed by the tail and time-to-crest), the lightning CFO and volt-time are governed primarily by the wave front (steepness), the tail being secondary except for very short, chopped waves. Breakdown voltage is high at short times and falls toward CFO at longer times, and the upturn at short times is steeper for more non-uniform gaps. A common approximation over about 2–11 µs is:

\[ V_B = \left(0.58 + \frac{1.39}{t}\right)\mathrm{CFO} \]

\(V_B\): breakdown crest voltage
\(t\): time to breakdown / flashover, µs

This general 2–11 µs fit gives \(V_B/\mathrm{CFO}\approx1.27\) at 2 µs and \(\approx1.04\) at 3 µs. Separately, suggested tower-insulation breakdown ratios are about \(1.67\,\mathrm{CFO}\) at 2 µs and \(1.38\,\mathrm{CFO}\) at 3 µs — these are a different reference from the curve above, not the same relation, so the two should not be mixed. The right volt-time curve depends on the configuration, polarity and waveform; general curves suit screening, but non-standard waves usually need a more specific model or validated data.

Section 10

Non-standard lightning impulses

Actual line overvoltages are often not 1.2/50 µs — shorter or longer tails, multiple peaks, oscillations and different fronts. For these, experimental data cannot be generalised; the response depends on the actual wave and the specific geometry. A simplified relation for the effect of the tail time on the CFO is:

\[ \mathrm{CFO}_{NS} = \left(0.977 + \frac{2.82}{t_h}\right)\mathrm{CFO}_{S} \]

\(\mathrm{CFO}_{NS}\): non-standard CFO for tail time \(t_h\)
\(\mathrm{CFO}_{S}\): standard 1.2/50 µs CFO
\(t_h\): tail (time-to-half) value, µs

Valid for times-to-crest of 0.5–5 µs and tails of 10–100 µs; \(\mathrm{CFO}_{NS}/\mathrm{CFO}_{S}\) rises as the tail shortens (a short tail keeps the voltage high for less time, so more voltage is needed to flash over). For the standard 50 µs tail the formula gives \(0.977+2.82/50\approx1.03\); some editions quote ≈1.016, so confirm the exact constant against your source. The practical rule: empirical CFO/BIL data suffice for standard impulses, but non-standard waves call for volt-time or leader-type methods (see the air-gap leader model) rather than a single BIL.

Section 11

Inputs for a lightning-performance study

A detailed line lightning-performance study needs shield-wire and phase-conductor models (several spans each side of the strike), frequency-dependent line sections, a tower surge model, the footing/grounding impedance, the power-frequency voltage at the instant of strike, the stroke current waveform and parameters, the stroke location, an insulation-strength model, and the statistical variation of lightning parameters. The stroke is usually an ideal current source defined by peak current \(I_p\), front time \(t_f=1.67\,(t_{90}-t_{30})\) and time-to-half \(t_h\), with these treated as lognormal random variables and the phase angle uniform. The objective is typically the backflashover rate, the shielding-failure rate, the total flashover rate, or the overvoltage distribution. APS develops these in the notes on lightning current waveforms, tower modelling, footing impedance and the CIGRE backflashover method.

Section 12

Power-frequency strength and contamination

For clean, wet insulation, power-frequency voltage rarely controls EHV/UHV line design — insulation chosen for switching and lightning is usually more than adequate at power frequency. The exception is contamination. Pollution, salt, industrial or agricultural deposits accumulate on the insulator surface; when the surface wets (fog, dew, light rain), the layer becomes conductive, leakage current flows, dry bands form and arc locally, and these dry-band arcs elongate and bridge the surface to flashover — at normal operating voltage, with no transient at all. Dry deposits do not reduce strength; it is the combination of contaminant and moisture that does. Contaminated lines tend to have repeated outages until the surface is cleaned, naturally or artificially. Rain itself can reduce the power-frequency strength of vertical (I-string) insulators by up to about 30%, while its effect on a pure air gap is negligible.

Contamination is a creepage problem

Mitigation is about surface, not air clearance: increase creepage distance, use fog-type, composite or silicone-rubber insulators, apply silicone grease, wash or clean, use semiconducting glaze, and base the design on site-specific pollution severity. Where switching overvoltages are well controlled on EHV/UHV lines, contamination can become the factor that sets the conductor-to-tower clearance.

Section 13

Power-frequency CFO of a clean clearance

For reference, a clean-air rod-plane power-frequency CFO (peak) is:

\[ \mathrm{CFO}_{PF,RP} = 750\sqrt{2}\,\ln(1 + 0.55\,S) \qquad\quad \mathrm{CFO}_{PF} = \mathrm{CFO}_{PF,RP}\,(1.35\,k_g - 0.35\,k_g^{2}) \]

\(\mathrm{CFO}_{PF,RP}\): rod-plane power-frequency CFO, kV peak
\(\mathrm{CFO}_{PF}\): power-frequency CFO with gap factor (gaps >2 m)
\(S\): strike distance, m
\(k_g\): gap factor

The peak power-frequency CFO is of the same order as — and for moderate gaps somewhat higher than — the minimum positive switching CFO, with the ratio increasing as the gap lengthens. The gap-factor form \((1.35\,k_g-0.35\,k_g^{2})\) returns \(\mathrm{CFO}_{PF,RP}\) at \(k_g=1\) and applies for gaps over 2 m; with it, the power-frequency CFO can exceed the positive switching CFO for ~3 m gaps with \(k_g>1\). But these clean-air values are reference only — for real lines, contaminated wet performance, not clean power-frequency CFO, is the controlling power-frequency consideration.

Section 14

Atmospheric effects

Flashover strength depends on air density and humidity. BIL and BSL are specified at standard atmosphere (absolute humidity 11 g/m³, 760 torr, 293 K). To translate a strength between conditions:

\[ V_A = \delta^{m}\,H_c^{w}\,V_S \]

\(V_A,\ V_S\): flashover / withstand voltage at actual and standard conditions
\(\delta\): relative air density
\(H_c\): humidity correction factor (\(H_c=1\) for wet/rain)
\(m,\ w\): air-density and humidity exponents (Table 2)

Generally flashover voltage rises with air density or humidity. The relative air density is set by pressure and temperature, and falls with altitude:

\[ \delta = \frac{P\,T_0}{P_0\,T} \qquad\qquad \delta = e^{-A/8.6} \]

\(P,\ T\): actual pressure and absolute temperature (K)
\(P_0,\ T_0\): standard pressure and temperature
\(A\): altitude, km

The altitude form \(\delta=e^{-A/8.6}\) (\(A\) in km) gives the mean relative air density; \(\delta<1\) above sea level, so air-gap strength falls with altitude. Humidity raises positive pre-breakdown strength and leader velocity but has little effect on negative lightning flashover; for wet/rain, \(H_c=1\).

Section 15

IEC altitude correction

IEC 60071-2 expresses the correction through a single factor and selects the exponent from \(G_0\):

\[ K_a = e^{-m\,A/8.15} \qquad\qquad G_0 = \frac{\mathrm{CFO}_S}{500\,S} \]

\(K_a\): atmospheric correction factor, \(V_A=K_a\,V_S\)
\(A\): altitude, km
\(m\): exponent (\(m=1\) for lightning impulse; switching depends on insulation type)
\(G_0\): parameter selecting \(m,\ w\) from Table 2

Direction matters. As written, \(K_a=e^{-m A/8.15}\) (\(A\) in km) gives \(K_a<1\) and converts a standard flashover/withstand voltage to its reduced at-altitude value (\(V_A=K_a V_S\)). The complementary IEC design-side factor — used to scale the required withstand up for a site above sea level — is the reciprocal, \(e^{+m H/8150}\) with \(H\) in metres; always state which direction is being applied. For lightning-impulse withstand \(m=1\); for switching it depends on the insulation type and coordination level. Correction beyond ~2000 m needs a modified procedure, and for UHV spacings the correction is best applied to the air-gap clearance rather than the voltage.

Table 2 — Atmospheric-correction exponents \(m\) (air density) and \(w\) (humidity) as a function of \(G_0=\mathrm{CFO}_S/(500\,S)\) (after Table 2.19, IEC/CIGRE).
\(G_0\) range	\(m\)	\(w\)
\(G_0<0.2\)	0	0
\(0.2<G_0<1.0\)	\(m=w=1.25\,G_0\,(G_0-0.2)\)
\(1.0<G_0<1.2\)	1	1
\(1.2<G_0<2.0\)	1	\(1.25\,(2.2-G_0)(2-G_0)\)
\(G_0>2.0\)	1	0

Section 16

BIL, BSL and CFO

Three terms are easily confused. BIL (basic lightning impulse insulation level) is a specified standard lightning-impulse withstand level. BSL (basic switching impulse insulation level) is the equivalent for standard switching impulses. CFO (critical flashover voltage, \(U_{50}\)) is the 50%-flashover voltage for a defined waveform, geometry and atmosphere. BIL and BSL are the withstand levels used in insulation coordination; CFO is a statistical flashover value. They are related — a withstand level sits a number of standard deviations below the CFO — but they are not the same quantity, and all three are referenced to standard atmosphere and corrected with the same factors.

Section 17

Practical assessment workflow

A practical external-insulation assessment runs:

Define the stress category — switching surge, lightning surge, clean power frequency or contaminated power frequency.
Define the insulation path — phase-to-ground, conductor-to-crossarm/tower/upper-structure, insulator string, or phase-to-phase.
Find the controlling strike distance and gap factor from the actual tower and conductor geometry.
Model the overvoltage source — simulate the switching event; for lightning, the stroke, line, tower and footing; for contamination, the pollution and wetting assumptions.
Select the strength model — a CFO expression, a BIL/BSL comparison, a volt-time curve, statistical strength or a standard design curve.
Apply corrections — atmospheric, altitude, wet/dry and pollution.
Compare overvoltage with corrected strength using the appropriate deterministic or statistical criterion.
Mitigate if short — increased clearance, longer or re-arranged strings, surge arresters, controlled switching, pre-insertion resistors, improved pollution performance or revised tower geometry.

Section 18

When simplified formulas are acceptable

Simplified CFO and gap-factor formulas suit early design estimates, screening of tower-window clearances, checking whether switching may control the design, rough comparison of configurations, understanding the effect of strike distance, and preparing inputs for insulation coordination.

When they are not enough

Do not use them blindly for final design when the tower geometry is unusual, phase-to-phase stress matters, the site is at high altitude, wet or contaminated conditions are critical, the waveform is strongly non-standard, the line is EHV/UHV with small margins, or statistical switching/lightning performance is required. Then use the applicable standard, test data, a validated design method or a detailed transient/statistical study.

Section 19

Main takeaway

External-insulation strength is not a single fixed number. It depends on the overvoltage type, waveform, polarity, strike distance, tower geometry, insulator arrangement, phase position, atmospheric condition and the statistical design criterion. Switching surges often control EHV/UHV external insulation because slow-front waves can fall near the critical time-to-crest of long clearances; lightning strength is assessed with CFO, BIL and volt-time behaviour, especially for the non-standard waves that travelling waves and tower response produce; clean power frequency rarely controls clearance, but contamination and wetting can control insulator performance. For preliminary work the simplified CFO and gap-factor formulas are enough; for final design, compare the actual simulated switching or lightning overvoltage against an insulation-strength method matched to the geometry, waveform, polarity and environment.

Suggested report wording

“The external-insulation assessment treated switching, lightning and power-frequency stresses as separate design conditions. Switching overvoltages were taken from transient simulations and compared with the applicable switching-surge strength (strike distance, gap factor, polarity, wet/dry condition and atmospheric correction). Lightning strength used the relevant CFO, BIL or volt-time characteristic; where lightning overvoltages were non-standard, waveform shape and time-to-breakdown were considered rather than a single 1.2/50 µs value. Clean power frequency was not taken to control clearance, but contaminated, wet insulator performance was considered where relevant. Atmospheric and altitude correction was applied where required.”

References

The standards, technical brochures, key papers and reference works behind this page.

IEC 60071-1:2019, Insulation Co-ordination – Part 1: Definitions, Principles and Rules. Geneva, Switzerland: International Electrotechnical Commission, 2019.
IEC 60071-2:2023, Insulation Co-ordination – Part 2: Application Guidelines. Geneva, Switzerland: International Electrotechnical Commission, 2023.
IEC 60060-1:2010, High-Voltage Test Techniques – Part 1: General Definitions and Test Requirements. Geneva, Switzerland: International Electrotechnical Commission, 2010.
CIGRE Working Group 33.07, Guidelines for the Evaluation of the Dielectric Strength of External Insulation, Technical Brochure 72. Paris, France: CIGRE, 1992.
A. R. Hileman, Insulation Coordination for Power Systems. New York, NY, USA: Marcel Dekker, 1999.
J. A. Martinez-Velasco, Ed., Power System Transients: Parameter Determination. Boca Raton, FL, USA: CRC Press, 2010.
IEEE Std 1313.2-1999, IEEE Guide for the Application of Insulation Coordination. New York, NY, USA: IEEE, 1999.

Part 16 Reading now

External-Insulation Dielectric Strength

Switching, lightning and power-frequency strength, gap factor and CFO, contamination and atmospheric correction.

Series progress 16 of 16

Part 15

Switching-Surge Insulation Strength

The active front, U-curve, gap factor and CFO formulas, and the statistical strength-vs-overvoltage design check.