Insulation Coordination

Beyond the Protected Section, and Design Observations

A protected line section is not a sealed box: the surge does not stop at its boundary. This closing part shows how a protected section can push high travelling-wave voltages into the first unprotected towers — moving the flashover rather than removing it — how far arresters must extend past the boundary in each stroke case, the counterintuitive way higher footing resistance can reduce stress, and the practical design observations on economics, tower spacing, shielded versus unshielded lines, and the value of a neutral.

Reading time ≈ 25 min

Section 1

What Happens Just After the Protected Section Ends?

Part Two asked whether arresters on some towers protect their neighbours. This closing part asks a sharper question: what happens just beyond the protected section? Suppose one section has arresters because the towers have high footing resistance, and the next section has low footing resistance, so no arresters are fitted. That sounds logical — but it can create a new problem.

Flashover can be moved, not removed

Flashover may be eliminated inside the protected section but shifted to the first unprotected towers just outside it — the protected section can “send” high travelling-wave voltages into the unprotected section.

This is Part Three, the final part of the line arrester series, building on the application, energy and intermediate-tower topics of Parts One and Two.

What this page teaches

why a protected section has no infinite protective range;
how high a voltage appears at the first unprotected tower, in each stroke case;
why extending arresters one or two towers beyond the boundary often fixes it;
why higher footing resistance can sometimes reduce insulation stress;
why the design starts with a performance target and economics, not equations;
why skipping towers raises energy and leaves gaps;
why shielded lines are easy and unshielded lines are demanding;
why a neutral makes low-voltage lines far more favourable.

Section 2

“Beyond the Protected Section”

Imagine the line split into a protected section (TWR1–TWR3 with arresters) and an unprotected section (TWR4–TWR5, none). A stroke inside or at the end of the protected section is clamped locally, so flashover may not occur there — but the travelling surge continues into the next towers, which have no arresters, so the voltage appears across their insulation.

No infinite protective range

Arresters protect locally and over a limited distance — they do not protect all towers beyond the protected zone. When arrester application stops, the first few towers after the boundary must be checked carefully, and sometimes arresters must be added on one or more low-resistance towers in the “unprotected” section.

Protected-section boundary check

check the first unprotected tower;
check at least the next few towers beyond the protected section;
compare the insulator voltage against the CFO;
extend the arrester section if the first unprotected towers become controlling;
do not assume that protecting the problem tower alone removes the line-outage risk.

Section 3

Stroke to Conductor, No Ground Wire or Neutral

This is the most severe boundary case: lightning strikes the phase conductor, there is no ground wire or neutral, the last protected tower has arresters, and the towers beyond it do not. With no shielding and no coupling benefit, the arrester clamps the voltage at the protected tower, but the wave still travels into the unprotected section and appears across its insulation. For a 115 kV line (\(R_1 = 29\ \Omega\), \(30\ \text{kA}\) stroke), the voltage at TWR2 and TWR3 reaches about \(1000\ \text{kV}\) — a real concern. The fix:

Table 1 — How adding an arrester at TWR2 lowers the unprotected-tower voltage without a ground wire.

Configuration	Voltage at the Unprotected Tower
No arrester beyond the boundary	\(\approx 1000\ \text{kV}\)
Arrester added at TWR2 (\(R = 10\ \Omega\))	\(\approx 558\ \text{kV}\)

Extend the arresters, not the whole section

The solution is not to arrest the entire next section — just extend the installation by one or more towers beyond the boundary.

Section 4

Low-Voltage Example: CFO Decides Acceptability

For a low-voltage line and a \(30\ \text{kA}\) stroke, the voltage at TWR1/TWR2 is \(790\ \text{kV}\); adding an arrester at TWR2 (\(R = 10\ \Omega\)) reduces the TWR3 voltage to \(328\ \text{kV}\). Whether that is acceptable depends entirely on the CFO:

\[ E_{\text{insulation}} < \text{CFO} \Rightarrow \text{acceptable}, \qquad E_{\text{insulation}} > \text{CFO} \Rightarrow \text{flashover risk} \]

Table 2 — Whether 328 kV flashes over against steel, concrete or wood pole insulation.

Voltage	CFO	Result
\(328\ \text{kV}\)	\(170\ \text{kV}\) (steel / concrete)	Flashover likely
\(328\ \text{kV}\)	\(1017\ \text{kV}\) (wood)	Flashover unlikely

Why wood performs better

The same \(328\ \text{kV}\) is unacceptable against a \(170\ \text{kV}\) CFO but safe against a \(1017\ \text{kV}\) CFO — which is why wood poles outperform steel or concrete: they add effective insulation strength.

Section 5

Stroke to Conductor, with Ground Wire or Neutral

With a ground wire or neutral grounded at the next tower, the case is less severe because of coupling. The analysis tracks \(e_g\) (ground-wire / neutral voltage) and \(e_c\) (phase voltage); the insulation sees only the difference:

\[ E_i = e_c - e_g \]

The insulation sees a difference, not the full voltage

The insulation does not see the full conductor voltage — only the difference between conductor and ground-wire / neutral voltage. If the ground wire / neutral rises with the conductor, the insulation stress is reduced.

Section 6

A Counterintuitive Result: Higher R₂ Can Reduce Stress

The text notes something that looks strange: as the footing resistance \(R_2\) at the unprotected tower increases, the voltage across the insulation can decrease. At power frequency, lower resistance is usually better — but for travelling waves the reflections depend on surge impedance and footing resistance, and a higher \(R_2\) can keep the ground-wire / neutral voltage more elevated relative to the phase, reducing the difference across the insulation.

Travelling waves break the usual intuition

For lightning travelling waves, low footing resistance does not always mean lower insulation stress at every instant — which is exactly why travelling-wave analysis is needed.

Section 7

Worked Examples with a Ground Wire or Neutral

For the shielded 115 kV line (\(R_i = 29\ \Omega\), shielding-failure current \(12\ \text{kA}\), \(e_g = 250\ \text{kV}\)), the insulation voltage at TWR2 (\(R_2 = 10\ \Omega\)) is \(450\ \text{kV}\) — about five insulators:

Shielded 115 kV needs no extra arresters

For the shielded 115 kV case, the ground wire plus arresters give sufficient protection beyond the boundary — arresters are not required on the unprotected towers.

A low-voltage line with a neutral is harder. For a \(30\ \text{kA}\) stroke (\(e_g = 712\ \text{kV}\)), the insulation voltage at TWR2 (\(R_2 = 10\ \Omega\)) is \(396\ \text{kV}\) — fine against a \(1017\ \text{kV}\) CFO but not against \(170\ \text{kV}\). Adding arresters helps progressively:

Table 3 — How successive arresters reduce insulation voltage on a low-voltage line with a neutral.

Action	Resulting Voltage
No extra arrester	\(396\ \text{kV}\) at TWR2
Arrester at TWR2	\(182\ \text{kV}\) at TWR3 (still > \(170\ \text{kV}\) CFO)
Arresters at TWR2 and TWR3	\(\approx 46\ \text{kV}\) at TWR4 (well below \(170\ \text{kV}\))

Very low CFO may need every tower

Two towers beyond the boundary usually suffice — but if a stroke terminates on TWR3, the TWR4 voltage rises again. For very low-CFO lines the stronger conclusion is that arresters may be required on all towers.

Section 8

Stroke to the Ground Wire at the Boundary

When the stroke hits the ground wire at the last protected tower (TWR1), a large surge still travels into the unprotected section. For the 115 kV example (\(100\ \text{kA}\), \(R_i = 29\ \Omega\), \(R_2 = 10\ \Omega\)), the ground-wire voltage is \(e_g = 2350\ \text{kV}\) and the insulation voltage at TWR2 is \(1230\ \text{kV}\) (maximum \(1316\ \text{kV}\)) — very high. Installing an arrester at TWR2 transforms it:

Table 4 — How an arrester at TWR2 cuts insulation voltage after a ground-wire stroke.

Configuration	Insulation Voltage Beyond the Boundary
No arrester at TWR2	\(1230\ \text{kV}\) (max \(1316\ \text{kV}\))
Arrester at TWR2	\(314\ \text{kV}\) at TWR3 (max \(\approx 348\ \text{kV}\), \(\approx\) four insulators)

Add an arrester at the next low-resistance tower

Even for ground-wire strokes, arresters may need to be added at the next low-ground-resistance tower to prevent flashovers in the unprotected section.

Section 9

The Boundary Cases at a Glance

Table 5 — The four boundary stroke cases with their risk and main design conclusion.

Case	Risk Beyond the Protected Section	Main Conclusion
Conductor stroke, no ground wire / neutral	High	Add arresters beyond the protected section
Conductor stroke, with ground wire / neutral	Usually lower	Shielded 115 kV may need no extra arresters
Low-voltage line, low CFO	High	Arresters may be needed on all towers
Stroke to the ground wire	Can still be severe	Add an arrester at the next low-resistance tower

Section 10

Design Observations: Economics and Alternatives

The first step is not a calculation — it is defining the required line performance (zero outages? a reduced rate? how critical is the line? who pays?). Only then do technical studies set the arrester rating, number, location, energy duty and flashover rate, after which the economics are compared:

Define the performance target

Technical study (rating, number, location, energy, flashover rate)

Compare alternatives

Economic decision

Arresters are not the only solution. Other methods — raising the insulation level, adding or improving ground wires, reducing footing resistance, improving the shielding angle or line configuration — may be more economical. Except for arrester failures, arresters can reduce the outage rate essentially to zero, but the design should still compare options.

Section 11

Design Observations: Energy and Tower Spacing

The author is honest that the hand equations are not fully successful for practical energy work, especially for many arrester towers (e.g. 11). Hand equations are valuable for understanding, screening, first estimates and sanity-checking simulations — but the final energy assessment should use ATP / EMTP®.

Skipping towers cuts cost but hurts twice

Arresters on every second or third tower reduce cost but (1) raise arrester energy — longer spacing means less energy sharing, so the struck arrester absorbs more — and (2) leave intermediate towers exposed, since a tower without arresters can flash over before the adjacent arrester's reflection arrives.

And, as Section 8 showed, if arresters cover only part of the line, flashover can appear just outside the protected part — so arresters should generally be applied to one or more towers into the unprotected section, rather than simply moving the flashover point to the boundary.

Section 12

Design Observations: Partial-Protection Strategies

Table 6 — Partial-protection strategies and their effect on the flashover rate.

Strategy	Effect
Arresters on the most exposed phase only	(outer phases of a horizontal line, lower phases of a vertical double-circuit) reduces the flashover rate but not to zero — partial protection
Arresters on one circuit of a double-circuit line	can eliminate double-circuit outages, but the other circuit may still flash over
Distribution / low-voltage, no ground wire	arresters are directly exposed to natural lightning — a more hostile environment than shielded transmission lines

Challenging but proven

Distribution arresters on unshielded lines face a hostile environment, yet field experience shows they have successfully protected equipment and survived — technically challenging, but possible.

Section 13

Specific Observations: Shielded Lines

For shielded lines the verdict is positive — generally no major technical problems. The ground wire intercepts many strokes, shielding-failure currents are limited (usually about \(5\)–\(10\ \text{kA}\) depending on insulation level), arrester energy is usually within capability, and the ground-wire-plus-arrester combination gives excellent protection.

Shielded lines are well suited

In some shielded cases arresters can be placed at every second tower, and — depending on insulation level — the adjacent unprotected section may need no arresters at all.

Section 14

Specific Observations: Unshielded Lines

For unshielded lines the outcome depends heavily on insulation level, since there is no ground wire to intercept strokes and the arrester is more directly exposed. For the 115 kV example, every-second-tower application and unprotected adjacent sections may be possible, but the arrester failure risk must be assessed probabilistically — the example failure rate is about \(0.55\%\) per year (energy withstand is not deterministic; it depends on lightning-current probability). Low-voltage unshielded lines hinge on the neutral:

Table 7 — How a neutral changes arrester spacing options on low-voltage unshielded lines.

Low-Voltage Unshielded Line	Outcome
Without neutral	Every-other-tower does not appear possible; adjacent sections may be unprotected; arresters may be required on the entire line
With neutral	The neutral acts as a coupling conductor (rises with the phase, reducing insulation stress) — every-second-tower and protected adjacent sections become possible in some cases

Section 15

Final Practical Conclusions

Table 8 — The final practical conclusions on applying line arresters beyond the protected section.

Conclusion
A protected section can push high voltages into the unprotected section
Flashover may be moved rather than eliminated
Arresters often need to be installed on one or more towers beyond the protected section
Shielded lines are generally suitable for line arrester application
Unshielded lines are much more demanding
Distribution lines without a neutral are especially difficult; with a neutral they are more favourable
Skipping towers cuts cost but can raise arrester duty and leave intermediate towers exposed
Hand equations aid understanding, but EMTP® is preferred for realistic energy studies

Line arrester application is a system-design problem — a practical checklist of the issues and their checks:

Table 9 — A design checklist pairing each issue with its practical verification check.

Design Issue	Practical Check
Protected-section length	Check the first unprotected towers beyond the boundary
Every-tower vs alternate-tower arresters	Verify the intermediate-tower CFO margin
Shielded line	Check backflashover and shielding-failure duty
Unshielded line	Check direct-stroke energy and failure probability
Poor footing resistance	Check tower voltage, arrester energy and reflections
Distribution neutral	Check neutral grounding and surge-current paths
Economics	Compare outage reduction against arrester quantity and maintenance risk

Section 16

The Simplest Way to Remember This

The surge does not stop at the boundary Think of the protected section as a safe zone — but the surge does not stop at the safe-zone boundary. It keeps travelling. If the next tower has no arrester, it can become the new weak point.

Equation Summary

Insulation Stress

\(\displaystyle E_{\text{ins}} = E_{\text{phase}} - E_{\text{gnd}}\)

The insulation sees the phase-to-ground-wire/neutral difference.

No-Flashover Criterion

\(\displaystyle E_{\text{ins}} < \text{CFO}\)

Insulation is safe while its voltage stays below the CFO.

Flashover Criterion

\(\displaystyle E_{\text{ins}} > \text{CFO}\)

Flashover occurs once the insulation voltage exceeds the CFO.

Coupling Factor

\(\displaystyle C = \frac{Z_m}{Z_c}\)

How closely the ground wire/neutral follows the phase voltage.

Boundary Insulation Voltage

\(\displaystyle E_i = e_c - e_g\)

Instantaneous insulation voltage from the conductor and ground-wire waves.

Maximum Boundary Voltage

\(\displaystyle E_{i,\text{max}} = e_c - C\,e_g\)

Worst-case boundary stress, eased by the coupling factor.

Arrester Energy

\(\displaystyle W_A = \int v_A(t)\,i_A(t)\,dt\)

Energy is the integral of arrester voltage times current over time.

Approximate Energy Estimate

\(\displaystyle W_A \approx E_A\,I_A\,T\)

A quick estimate from discharge voltage, current and duration.

Memory map. Clamp at the protected tower → the surge travels on → check the first unprotected towers → compare \(E_{\text{insulation}}\) with CFO → if too high, extend arresters one or two towers into the “unprotected” section → on very low-CFO lines, consider all towers.

Final engineering message

The design question is not only “are the protected towers safe?” but also “what happens to the first few towers after the protected section?” A protected section can send high travelling-wave voltages into the unprotected section, moving the flashover rather than removing it — so arresters often need to extend one or more towers past the boundary. Shielded lines are generally well suited; unshielded lines, and especially low-voltage lines without a neutral, are far more demanding. And while hand equations build understanding, realistic arrester energy assessment belongs in EMTP®.

Part Three Reading now

Beyond the Protected Section and Design Observations

Why a protected section pushes high voltages into the first unprotected towers, how far to extend arresters past the boundary, and design observations on economics, tower spacing, shielded versus unshielded lines and the neutral.

Series progress 3 of 3

Part Two

Arrester Current, Energy and Intermediate-Tower Protection

Estimating arrester current and energy with travelling-wave and inductance analysis, why arrester current is not the stroke current, energy sharing across towers, and whether every-second-tower arresters protect the intermediate tower.