How placement works

A layout is built in one pass, in a fixed order. Each stage narrows the ground the next stage is allowed to use, so the order is the design: the boundary becomes a usable area, the usable area is filled with module tables, and every piece of equipment placed after that takes tables back out again.

Knowing the order tells you which input to change when a result is not what you expected. A plant that came out smaller than the site area suggests has usually lost ground at one specific stage — the perimeter road, a transmission line corridor, excluded terrain, or the footprints of the control rooms and arresters — and each of those is a different field in the input panel.

The usable area

Usable area is the ground a module table is allowed to stand on. The plant boundary is shrunk inward by the perimeter road width to give the usable area; obstructions and terrain exclusions are then subtracted.

In full, the usable area is what remains of the boundary after four subtractions:

Everything placed later is checked against that one area. Tables, control rooms and lightning arresters must sit inside it, and every candidate cable route is validated against it before it is used.

Cables are the single exception to the road band. Cables may run inside the perimeter road band, but never outside the plant fence — the road is a service corridor, not an electrical exclusion.

The road band itself is computed once, as the strip between the boundary and the shrunk boundary, and the same strip is used by the on-screen plot, the PDF report, the Google Earth file and the CAD drawing. The road you see on screen is the road in every export.

The stages in order

1. The boundary file is read

The application reads the plant boundaries out of the file, and with each boundary the obstacles and line features that lie inside it. A polygon that is not contained within another polygon is a plant boundary; a polygon fully contained within a boundary is an obstacle. Which kind of obstacle it is comes from its name, which is why the naming rules in Boundary file requirements decide whether a pond is treated as water or as a plain obstruction.

Several boundaries in one file are laid out as several plants. See Multi-plot sites.

2. Everything is converted to metres

Latitude and longitude are projected into the site's local UTM zone, and all internal geometry from this point on is in metres. Conversion back to latitude and longitude happens only at the edges — when imagery is drawn behind the layout, and when a Google Earth file is written.

This is why every dimension you enter is a length in metres and never an angle or a coordinate: gaps, pitches, footprints, setbacks and radii are all measured on flat ground in the site's own zone.

3. The boundary is shrunk inward by the perimeter road width

The boundary is offset inward by the Perimeter road width — 6.0 m by default — all the way round. What is inside the offset is available for building; the band between the offset and the fence is the perimeter road.

Set the width to 0 and no band is reserved, and tables are then allowed to reach the fence line itself.

4. Obstructions and corridors are subtracted

Every polygon obstacle read from the file is cut out of the area: water bodies, buildings, substations, and any obstruction you draw by hand. Line features are different — a transmission line, canal or road has no width of its own, so each is given a setback on each side of the line. The corridor cut out of the site is therefore twice the setback: at the default Transmission line corridor of 15.0 m per side, a 30 m corridor.

Raise the setback for a higher-voltage line. The application's own tooltip suggests 30 m per side for a 400 kV line.

5. Steep and unsuitable ground is excluded

When Avoid steep / unsuitable ground is on, an elevation model of the site is built — from a contour file you supply, or from public satellite elevation data — and ground that fails your slope limits is cut out of the usable area as well. Ground relief within a single table footprint is limited separately, as the pile-reveal differential you are prepared to build.

Terrain is the one stage that can fail harmlessly. If there is no contour file and no internet connection, the layout proceeds with no terrain exclusion rather than stopping.

6. The table grid is placed

The grid is laid across whatever ground is left. Rows run east to west; panels face the equator — south in the northern hemisphere, north in the southern.

7. Only whole tables are kept

A table or tracker unit is placed only where it fits entirely inside the usable area. A position where any part of the footprint crosses the road band, an obstruction, a corridor or excluded terrain is skipped — nothing is clipped, trimmed or half-built.

This rule is what leaves the ragged edge along a diagonal or curved fence: the grid is regular, the boundary is not, and the last position in each row is rejected as soon as one corner falls outside. Two options exist to win that ground back — see Maximizing placement and half tables.

8. Control rooms are placed, and the count re-checked

The plant's capacity is now known, so the number of inverter control rooms (ICRs) follows from it: one room per ICR Block of capacity, 18.0 MWp by default, rounded up. The tables are grouped spatially by position into blocks of roughly one block-size each, and a room is placed at the centre of each group, so every group of tables surrounds its own room.

Each room must fit entirely inside the usable area, and any table overlapping a room's footprint is removed. That drops the plant's capacity — which is why the number of rooms is then re-checked against the new, lower capacity. See Inverter control rooms.

9. Inverters are placed and cables routed

Inverters — or string monitoring boxes (SMBs) in Central Inverter mode — are placed from groups of tables, sited in the gap bands between rows rather than in the middle of one. A placed string inverter is drawn as a small marker, 2 m east to west by 1 m north to south. Their counts are reported whether or not cables are routed.

Routing the cables is optional, and it is the slow step on a large or complex plant. With Calculate Cables for PV Power Plant off, the counts still appear and every cable length column reads —. With it on, DC runs, AC runs and the medium-voltage runs between control rooms are routed and measured. On a very large plant the application switches to fast geometric estimation so generation stays quick. See Cable routing.

10. Lightning arresters are placed, and capacity recomputed

With Place Lightning Arresters on, arresters (LAs) are laid on a coverage grid whose spacing is the protection radius you set, so neighbouring protection circles overlap and no part of the plant is further than one radius from an arrester. On a fixed-tilt plant an arrester has a real footprint and the tables it covers are removed, so capacity falls once more and the plant totals are recomputed. On a tracker plant the arrester is a slim pole placed in the gap between tracker columns, and no units are lost. See Lightning arresters.

With the option off, no arresters are placed at all and tables fill the whole usable area. You can still add them by hand later in Sketch Mode.

Why the order matters

Three of the stages remove tables that stage 6 had already placed: the control rooms, the arresters, and any obstruction you add by hand. Capacity is therefore not final until the last of them has run.

That is the reason for the re-check in stage 8. The number of control rooms is derived from capacity, and capacity is lower after their footprints have been cleared than it was when the count was first worked out. Re-checking against the reduced figure is what stops the plant carrying a room it no longer has the capacity to justify.

The same logic explains the order of the equipment stages. Control rooms are placed before cables because the cables terminate at them; arresters come last because they are the only equipment whose position is set by coverage geometry rather than by the plant it serves.

Reading the plot

The plot fills the right-hand side of the window, with the toolbar above it and the Layout & Energy Summary table below.

The toolbar

Screenshot pendinglayout/toolbar.png

The plot toolbar

What it shows

The full toolbar row, wide enough that every button label is readable.

How to get there

After a layout has been generated, so the satellite and piles buttons are enabled.

🛰 Satellite is the fastest check that a layout is standing on the ground you meant. Real tracks, ponds and buildings appear behind the tables, so a missing obstruction in the boundary file is obvious.

Screenshot pendinglayout/satellite.png

Layout over aerial imagery

What it shows

The plot with the aerial photograph behind the layout, so the placement can be read against real ground features.

How to get there

Generate a layout, then switch the satellite button on. Needs an internet connection.

▢ Wireframe is the view to use when tables are dense enough to merge into a solid block of colour, and when you want to see cables or piles that filled tables would otherwise hide.

Screenshot pendinglayout/wireframe.png

Outline drawing

What it shows

The plot with tables drawn as outlines only, ideally at a zoom where individual tables are distinguishable.

How to get there

Generate a layout, then switch wireframe on.

View switches

Below the toolbar is a block of ON/OFF switches, each controlling one layer of the drawing.

Screenshot pendinglayout/view-toggles.png

View switches

What it shows

The whole block of switches with each label and its current state readable, at their default states.

How to get there

After a layout has been generated. Do not change any switch first.

The three cable layers and the arrester layer start off because a first run usually has neither cables nor arresters to draw. Switch them on after a run that produced them.

The colour key

The Legend switch draws the key on the plot itself. The colours are fixed:

Contour lines, when the Contour / Terrain switch is on, run green at the lowest elevation through blue to red at the highest, and the excluded area is reported in acres.

Fitting the view

The navigation toolbar under the plot is the standard set of pan and zoom controls, with one difference worth knowing: Home (⌂) fits the view to the plant rather than returning to some earlier zoom. Whatever you have panned or zoomed to, one click frames the whole plant again.

For anything you need to look at closely, ⛶ Expand Plot gives the drawing a window of its own; ⟵ Return to Main Window brings you back.

Screenshot pendinglayout/expand-plot.png

Expanded plot view

What it shows

The expanded plot window with the return button visible.

How to get there

Click Expand Plot on the toolbar.

Callouts to add

Outline the return button.

The summary table under the plot has its own ⛶ Maximize button, which is the only comfortable way to read the full column set. What each column means is in Summary table columns.

Where to go next