Distance between conductors: the core's part#

A design code may require a distance between two conductors of a span that depends on where along the span one looks — on the local sag, the suspension strings, the angle of the line between the conductors. OpenOHL measures (the conductors, their pairs, stations along them: the geometry); a check plugin holds the code's values and its rules and says what is required at each station, through the check interface (Writing plugins). A plugin for the Swiss LeV runs such a check; it is kept apart from the public code. The measuring is ohl_clearance::phase_distance, the host's side ohl_io::checks, the window crates/ohl-app/src/phase_distance.rs.

What is measured#

For each pair of conductors in each weather case chosen (each case checked on its own, both conductors always from the same case):

  1. Stations along the first conductor, every Δs of its length (default 1 m), its ends included.
  2. At each station P, the closest point Q on the second conductor (3D). The measurement is made in the vertical plane through P and Q (not square to the span: with large line angles a square section would compare points far apart along the other conductor).
  3. Distance d = |PQ|.
  4. Local sag of each conductor at its point: the vertical distance from its chord (the straight line between its clamps) down to the conductor — 0 at the clamps.
  5. l_k of each conductor at its point: the length of the suspension string, linear along its span between its two ends. At an end: a suspension I-string, its length; a V-string, a tension string or a wire clamped straight to the structure, 0; a V with a string below, that string's length; a TAK, its suspension string's length.
  6. β: the angle of PQ from the vertical.
  7. With phase spacers joining the two conductors in the span, each one's largest sag between its spacer clamps (each spacer taken as an attachment point: the sag of each stretch on either side of it below that stretch's chord, the biggest counting); the local sag stays the one between the span's clamps.

The stations are taken on each conductor of the pair in turn, so a short and a long conductor are both covered. For a bundle, each subconductor's stations are measured to the nearest subconductor of the other conductor.

The plugin receives, per pair, what its rules need besides the stations (conductor-pair): whether both are in one electrical system, whether they hang between the same two structures, each one's nominal voltage (none: an earth wire), the plugin columns of their sections, the mean conductor height of the span, the diameters and masses, the sags between spacers; and per station the six numbers d, β, each local sag, each l_k. It returns per station the required distance first, then its own values (kept with the worst station), and the rule applied.

Which pairs#

  • Conductors (sections' spans; bundles: the nearest subconductors) hanging between the same two structures.
  • Only the circuits ticked; a pair is checked when both its conductors' circuits are.
  • Nearby pairs (a tick box): pairs of conductors not between the same two structures (another line's spans, the next span's) whose boxes come within a maximum distance (default 20 m), measured only where closer than that.
  • Not paired: jumpers, two earth wires, two conductors of the same circuit and phase (one potential), two sharing a clamp.

Report#

Per pair: the worst station (its two points, d, the required distance, the margin, the measured values and the plugin's own, the case) and the length of conductor below the required distance (on the conductor where it is longest, each stretch counted once). Shown in the check's window, worst first, exported as CSV; a worst station at a clamp is marked as such. Clicking a result draws the line between its two points in the 3D view, labelled with d and the required distance. Runs are kept in the library of performed checks (Analyses).

The values#

The plugin ships its values file (TOML, its own extension) and shows it as tables (value-tables), locked in the window; a project admin may edit a cell (edit-value, checked by the plugin), kept as the project's copy and used by every run on the project and kept in each report.

Speed#

On an existing line (930 conductors, 5593 pairs between the same structures; about 10 500 with nearby pairs within 20 m): about 0.8 s on all cores, the closest point searched in boxes of 8 pieces, nearest first; the plugin's evaluation costs about as much as the same computation built into OpenOHL, plus about 0.3 s to load the plugin once.

View this page's source on GitHub