Decisions · Span screening

Which spans expose the middle phase?

The paper ends with a five-step procedure for a line with R/T arresters. For every span: find how high phase S stands above the ground beside the line, turn that into the largest current that can still reach it, compare it with the current that flashes it, add up the strokes, and decide where phase-S arresters are needed. Here it runs on all nine spans from T16 to T25. Click a tower to give it phase-S arresters.

  1. 1Ground and heightHeight of phase S along the span and the ground beside it.
  2. 2Capture surface, e, ImaxEqs. (4)–(6): the largest current whose leader can still reach S.
  3. 3Exposed lengthWhere Imax exceeds Ic = 2Uref/(kZc), eq. (7).
  4. 4Outage rateNS = 10⁻⁶·Ng·(d/π)·∫[P(Imax) − P(Ic)]⁺dx, eq. (8).
  5. 5ArrestersBoth ends, unless all the exposure lies within xp of one tower, eq. (9).
  6. ✓CheckResidual rate after the arresters you place.
Ic = 5.7 kA follows from eq. (7) with Uref = 912.5 kV, Zc = 535 Ω, Zch = 400 Ω. xp = c(τf − τg)/2 with τf = 0.8 µs and τg = 0.3 µs.

T16 → T25: maximum attachable current on phase S

click a tower to add or remove phase-S arresters · click a span for its profile
Imax ≤ Ic: no flashover-capable stroke can reach SImax from Ic to 18 kAcovered by phase-S arresterstower with phase-S arresters

Span profile

Imax C1 sideImax C2 sideIcreach of a phase-S arrester

Screening result

SpanLImax C1 / C2ExposedNSOne end?Residual

NS is shown as one flashover-capable stroke on phase S every so many years, both sides together. "One end?" names the tower whose arresters alone would cover all the exposure, if one exists.

What to take away

Four lessons

Clearance is the whole story

Imax depends only on how high phase S stands above the ground capture surface. Near tall towers and over valleys it is high; at the low point of a span over level ground it can drop below Ic, as in the middle of T18–T19.

The window is narrow but wide enough

Only strokes between Ic ≈ 5.7 kA and Imax ≈ 10–17 kA matter. That is a small slice of all strokes, so each span side sees one such stroke every one to several thousand years. Across a fleet of spans it adds up.

One arrester rarely covers a span

The relief wave of a single arrester protects only about 75 m from its tower. The exposure on these spans runs along their whole length, so one end is never enough here. That is the paper's result for T20–T21: arresters at one tower failed in all six tested cases.

Terrain is missing for eight spans

Only T20–T21 has a terrain survey. The others are shown over level ground. Use the slider under the profile to see how a valley or a hillside beside a span would raise its Imax. Surveying those spans is the next step before placing arresters.

Data and model
Heights: top phase R at the supports and at mid-span from the archived EMTP geometry of the nine spans (EGM_All_Nine_Spans), parabolic in between; phase S 4.7 m below R. T20–T21 with terrain: the 25 surveyed cross-sections and the 30 m sag of the paper, through the same model as the 3D Strike Lab, so the C1 side reaches about 17 kA (Brown–Whitehead) and 13 kA (Love) over the valley as in the paper. Level spans: Imax from eq. (6) with heff = hS + Δ. Ng is the mean of the two towers' local values (7.7–13.0 flashes/km²/yr). Residual: a point is covered if both towers have phase-S arresters or it lies within xp of an equipped tower.

A screening tool, two-dimensional, with assumed sag. It identifies spans to survey and protect; it does not replace the transient check of the Relief Wave Lab.