Dimensional Coordination & Grids

Start with the problem dimensional coordination solves. When you build on site, dimensions can stay a little loose because skilled trades reconcile them in place — a bricklayer adjusts the perpend joints across a wall, a carpenter scribes a skirting to an uneven floor, a plasterer floats a wall flat. The building absorbs imprecision through wet trades and craft. Take that craft away — manufacture the parts in a factory and merely assemble them on site — and there is nothing left to absorb imprecision. The parts must be designed to fit before they are made, because once they are made, they are made.

The way you guarantee fit in advance is to make every part obey the same underlying measurement system. This is dimensional coordination (sometimes modular coordination): a shared set of rules about sizes and positions so that components, panels and modules are standard and interchangeable — a part made to the system fits anywhere the system says it should, without cutting or packing. It is exactly the logic of interchangeable parts that made mass manufacturing possible in every other industry: a rifle part, a car part, a phone part is made to a specification, and any conforming part fits any conforming assembly. Buildings were slow to adopt this because site craft hid the need for it; off-site construction removes the hiding place.

The consequence is a reversal of habit. On a bespoke site-built job, you design the form you want and work out the dimensions afterwards. In off-site work you design to the dimensional system from the first sketch — rooms, openings, panels and modules are chosen from a coordinated menu of sizes rather than invented freely. This feels like a constraint, and it is; but it is the same constraint that lets a factory hold quality and drive down cost through repetition. A designer who treats the grid as a nuisance to be escaped will fight the system the whole way. A designer who treats it as the enabling structure — the reason a good part can be repeated — works with the grain of off-site construction. The rest of this lesson is about the two tools that carry that discipline: the module of measurement, and the grid it is drawn on.

The first tool is a basic module: a single small unit of length that every coordinated dimension is a whole multiple of. Internationally this basic module is conventionally written M, and in much of the world M = 100 mm — a convenient unit because it divides and combines cleanly and suits the scale of building components. (The exact module is a convention of the system you are using; treat the number as illustrative, not a rule.) Once you have M, you build larger sizes as multiples: 3M, 6M, 12M and so on. Out of these, systems pick a shorter list of preferred dimensions — the sizes that will actually be used for rooms, grids, panels and modules — so the menu stays small and the parts stay few.

Why bother? Because a small, disciplined menu of sizes is what makes parts interchangeable and re-usable. If every wall panel is a preferred width, a panel made for one place fits many places; stock, jigs and details are shared; an error is corrected once across the whole set. The moment you allow an off-module size — a room that is 2.6M wide because it looked nice in plan — you have created a special: a one-off part that needs its own jig, its own drawing, its own handling, and that breaks the interchangeability for everything around it. A few specials are fine and sometimes necessary; a building full of them has thrown away the economics of off-site altogether.

There is a hierarchy worth naming. Controlling dimensions are the big coordinating sizes — structural grid spacings, floor-to-floor heights — that everything else works within. Within them sit the coordinating sizes of components (the space a panel or module is allocated, including its joints), and within those the work sizes (the actual manufactured size of the part, deliberately a little smaller than its coordinating size to leave room for the joint — a distinction we develop properly in lesson 5.2 on tolerance). The key idea to carry now is simple: choose sizes from a coordinated module and its preferred multiples, not freely. Standardisation is not the enemy of good design; it is the discipline that lets a few well-made parts do the work of many, and the foundation on which everything else in off-site construction is built.

The second tool is the grid: a set of reference lines, spaced on the module, that everything is positioned against. In practice a building carries more than one grid, and the art is in making them agree. The planning grid (or dimensional grid) governs the arrangement of spaces — room widths, corridor positions, the repeating bay of accommodation. The structural grid governs where the load-bearing elements sit — columns, load-bearing walls, the lines the floor spans between. On a well-coordinated off-site project these are not two independent drawings fighting each other; they are locked together so that structure lands on planning lines and modules sit cleanly within structural bays.

There is a subtlety about *how* a grid line relates to the thing it positions. A line can be an axial reference (runs down the centre of a column or wall) or a face reference (runs along a face). Mixing the two carelessly is a classic source of dimensional confusion: a nominal grid dimension that forgets whether it is measuring centre-to-centre or face-to-face will be out by the thickness of a wall every bay. Off-site work demands you are explicit and consistent about the convention, because the factory will build exactly to whatever the drawings say, wall thickness and all. The grid also needs to decide where the joint lives relative to the line — whether the module face sits on the grid or is held off it by a designed gap — which again points forward to tolerance.

Think of the grid as the score that the whole orchestra plays from. The structural engineer, the module manufacturer, the services designer and the interior designer are all reading positions off the same lines; when the lines agree, their work meets cleanly on site. When two of them are secretly working to slightly different grids — one to column centres, one to module faces — the mismatch surfaces at assembly, when it is most expensive to fix. So the grid is set early, agreed by the whole team, and treated as close to sacred. Its spacing is not arbitrary either: as the next section shows, the planning grid is pulled one way by the rooms and the other way by the truck, and a good grid is the number that satisfies both.

Here is where dimensional coordination stops being abstract and becomes a design decision with teeth. In volumetric off-site construction, the module is a box that must be made in a factory, driven down a public road, and lifted into a building — and the grid has to reconcile all three. The building wants a grid that makes good rooms. The road imposes a hard ceiling on how wide (and long and tall) the box can be before it needs special permits or becomes un-transportable — a limit set by transport regulations and the route, not by the designer. The factory wants a size its jigs and bays are built around. And the crane wants a weight and reach it can manage. The planning grid is the number where all of these meet.

That is why you so often see the same coordinating width appear across very different modular buildings: it is roughly the widest good room that still fits within ordinary road-transport limits. Push the module wider for a more generous room and you may cross into oversize-load territory — escorts, permits, night runs, route surveys, cost and risk — or off the legal envelope entirely. Shrink it and rooms feel mean and you carry more joints per metre of building. The grid is the negotiated settlement: rooms on one side, the truck on the other. (The actual dimensions are set by the chosen system and the governing transport rules of the region — treated in Module 7 — and are never a number to assume.)

Once the grid is fixed, it propagates. The module width sets the planning bay; the planning bay sets the structural grid; the structural grid sets the foundation setting-out; the same grid governs where services risers land and where the interior partitions fall. A single coordinated grid is what lets the foundations be poured on site (to that grid) while the modules are still being built in the factory (to that same grid) — the parallel working that gives off-site its speed only works because both halves are measuring to the same lines. This is the deepest reason dimensional coordination is the foundation of off-site construction: it is the common language that lets the factory, the road, the crane and the site all build one building without ever being in the same place at the same time. Get the grid right and everything downstream has a chance. Get it wrong and no amount of good detailing later will rescue it.