Station pointing

Station pointing displays the aircraft as seen from below and lets you click directly on the image to position its stores stations. It has its own screen in the Parts & breakdown menu, called Station pointing, and it is what makes everything that follows possible: a read-only plan on the aircraft record, hooking up straight from the drawing, comparison thumbnails.

Pointing never creates a station: it positions the ones the applicable configuration already declares (“Stores station” nature), together with their compatibilities.

Which configurations are listed

Only the configurations that carry stores: those whose target (family, variant or version) belongs to a family of a category checked “Carries stores” (see the Family categories screen). A part configuration — an engine, an avionics bay — never carries any: an engine has no hardpoint.

Each row tells how far pointing has gone:

Stations — how many are pointed out of how many are declared. “No station” as long as the configuration declares none.
Underside view — whether the aircraft image is uploaded and calibrated. A family or variant configuration can use several versions as a background: the column tells how many are calibrated.

“Point” opens the full-page plan. From an applicable configuration that carries stores, the “Point the stations” link leads there too.

A rule enforced by the server

The “Carries stores” box is not a display filter, it is a rule:

Employment point nature. The “Stores station” choice is only offered on a configuration that carries stores. Anywhere else, a station or a position is refused, with a message naming the cause.
Retargeting. A configuration that has stations cannot be pointed at a target that carries none.
Categories and families. A category cannot be unchecked, nor a family moved to an unchecked category, while stations depend on it — across all instances, since the product referential is shared. Remove those stations first.

Why metres, not pixels

What gets saved is never a position on an image — it is a position on the aircraft. Concretely: the screen converts every click into metres, in the aircraft’s own frame (origin at the nose tip, abscissa aft, signed lateral offset), and that is the coordinate that gets stored. The image is only a calibration backdrop — a support to click at the right spot, nothing more.

Direct consequence, and the whole point of this choice: changing the image never destroys a position already placed. You can replace a rough first scan with a clean technical drawing, or pick a different product version as the backdrop, without ever having to re-point a single station. Had the screen stored pixels instead, the data would be tied to the file — the day it changes, everything would need redoing.

The orientation convention

Aircraft frame: Origin at the nose tip, on the centreline. Abscissa (x) aft. Lateral offset (y) signed, positive to starboard. Every station coordinate is expressed in this frame, in metres.

The plan is a view from below: the aircraft seen from the ground, nose toward the top of the screen. In that convention, starboard appears on the READER’s LEFT — counter-intuitive the first time, and the single most common source of a mirrored placement. The screen restates it at every step of calibration.

Calibration: two clicks, then a third one that protects you

Before a single station can be positioned, the image must be calibrated: tied to the aircraft frame through two reference points, each known both in pixels (where you click) and in metres (what that point actually represents — typically the nose tip at 0, 0 and a point on the centreline at the overall length).

Calibrate the image

First reference point
Click an unambiguous, identifiable point (the nose tip, for instance) and enter its metric coordinates.
Second reference point
Click a second point — typically a point on the centreline at the overall length — and enter its coordinates. The derived scale (metres per pixel) shows immediately: the first plausibility check, a technician who roughly knows the aircraft’s size sees right away if the result is absurd.
Control point — the STARBOARD wingtip
Click the starboard wingtip. This point plays no part in computing the calibration — it only VERIFIES it (see below for why it is indispensable). The measured offset must come out positive; otherwise the calibration is refused and never saved through PATCH/product-versions/:id — which carries the underside image and its calibration.

Why this third click, and why it is counter-intuitive. Two points are mathematically enough to calibrate an image: they fix scale, rotation and position. But they never fix mirror orientation — and that is the trap. If the image is flipped (a top view served by mistake instead of a bottom view, or a file exported mirrored), calibrating on those same two points looks perfect: no error, no residual to measure, nothing flashing red. Everything placed afterwards simply comes out… inverted. A station meant to sit on the left shows up on the right, silently, with no calculation ever noticing. The only way to catch this is to check a third point whose expected sign is known in advance — hence the starboard wingtip, which must come out positive, no exceptions.

Placing a station

Once calibration is in place, select a stores station and click on the image: the metric position is computed immediately and stays displayed at all times, for checking against the aircraft’s published dimensions. The move is saved through PATCH/employment-points/:id — which carries the station coordinates (stationX, stationY).

To find a station on the drawing, hover over it in the list: its marker grows and lights up on the plan (keyboard focus on “Re-point” does the same). On a large screen, the plan stays in view while you scroll down a long list of stations.

The symmetry assistant never guesses

An explicit “Mirror to…” action copies one station’s position onto another, flipping the sign of the lateral offset — handy for a left/right pair. It only ever fires on request: the screen never assumes that one station is another’s mirror image. An assumed symmetry would be an invisible error, exactly like the flipped-calibration trap above — and a pair placed by hand is never perfectly symmetrical anyway.

An unpositioned station is not an anomaly

Every station that exists today has no coordinates yet: that is the normal starting state, not an urgent error to fix. The screen permanently shows how many stations remain to be positioned, and names them — never a silent absence.