Also searched as aluminium profile sizes, 20x20 to 100x100 aluminium profiles, aluminium profile size chart, t-slot profile dimensions, aluminium extrusion sizes guide, 40x40 45x45 60x60 profile comparison.

Two sections that are four per cent apart

Start with the pair that destroys the assumption behind most size charts. Take a 50 × 50 four-slot section and a 60 × 60 four-slot standard section, both in 6063-T5, both from the same extrusion house.

The 60 × 60 is a bigger member by every outside measurement, and it is barely stiffer than the 50 × 50. It weighs 39 per cent more, it costs more per metre, and it moves the whole frame into a heavier class of brackets, feet and end caps. What you are actually buying at 60 mm is not stiffness. It is the 10.2 mm slot mouth, the 14 × 14 mm internal groove, the 48 mm pitch and the larger centre bore — access, fastener size and cable routing.

The three numbers that decide, and the one that gets forgotten

Four things move the load a profile can carry, in this order of influence: the depth of the section, the wall thickness, how the material is distributed relative to the neutral axis, and which way round you install it.

Depth dominates because stiffness scales with roughly the fourth power of it, which is why stepping from a 30 to a 40 series is worth more than any wall change inside either. Wall thickness comes second and is invisible in the name — a “4040” is three different products. Material distribution is what separates a T-slot profile from a plain tube: the slot walls and the centre boss put metal where it resists bending, at the cost of a thinner outer flange.

Orientation is the one that gets forgotten, and it can be worth more than a size change. Rectangular and deep beam sections have two very different inertias, and the gap is much wider than most people expect.

A 30 × 120 installed the wrong way round is a 12 cm⁴ member — worse than a 45 × 45 standard — and it weighs 3.10 kg/m. Installed the right way round it is a 115 cm⁴ beam, better than a 60 × 60 heavy at half the weight. Nothing about the part changes. Only the drawing does.

What each size carries, at three spans

The table below is a single simply supported member carrying a point load at mid-span, with the deflection limited to L/300 — the usual limit for machinery, guarding and conveyors. The three load columns are allowable load in kilograms at a 1 m, 1.5 m and 2 m span. A factor of four on the span costs a factor of eight on the load, because deflection goes with the cube.

Two patterns fall out of the last column. The first is that efficiency falls as the member grows: a 20 × 20 gives 1.1 cm⁴ per kilogram per metre, a 40 × 40 standard gives 5.3, and a 60 × 60 heavy gives 5.9 for three and a half times the weight. The second is that the deep beam members break the pattern completely. A 30 × 120 on edge gives 37 cm⁴ per kilogram per metre — seven times the efficiency of a 40 × 40 standard — because depth, not material, is what buys stiffness.

What fails first at each size

Every step up the range has a characteristic failure mode. Knowing which one you are buying your way out of is more useful than knowing the load number, because the failure is what the customer sees.

The upgrade ladder, priced in stiffness per kilogram

Steps up the series are not equally valuable, and the difference is large enough to plan around. This is the same table as before, rearranged as a ladder.

The last row is the one worth remembering. Changing the shape of the member instead of the series buys 2.6 times the stiffness for 1.46 times the weight, and it keeps the same slot, the same hardware and the same end caps. The only cost is 20 mm of depth in one direction. Where that depth is available, it beats every series step on the list.

Four alternatives to going up a size

Before ordering a bigger section, work through these four. In practice at least one of them applies, and each is cheaper than the next series.

1. Add a support. Deflection goes with the cube of the span, so halving the unsupported length multiplies the allowable load by eight. Adding one post and two brackets to a 2 m span turns a 26 kg 40 × 40 standard member into the equivalent of a 208 kg one. No member change comes close to that return.
2. Turn the member through 90 degrees. A 40 × 60 gives 24 cm⁴ on edge and 13 cm⁴ flat. A 30 × 120 gives 115 against 12. Check which axis the load actually bends the member about, and check that the bracket you are using does not force the weak axis into the load path.
3. Change the support condition. The deflection coefficients for the four common cases are 3 for a cantilever, 48 for a simply supported span, 110 for one end fixed and 192 for both ends fixed. Converting a cantilever to a simply supported span is worth a factor of sixteen — more than any size in the catalogue.
4. Change the material distribution, not the envelope. A deep beam at 30 × 120 carries 115 cm⁴ at 3.10 kilograms per metre. A 60 × 60 heavy carries 38 cm⁴ at 6.40 kilograms per metre. Where the load is in one plane and the depth is free in one direction, the deep beam is several times more efficient.

The pattern is consistent: geometry beats material, and support layout beats geometry. A section change is the most expensive way to buy stiffness, and it is the one that most projects reach for first.

Tolerance loosens as the section grows

One consequence of going up a size that nobody mentions until the frame is being assembled: the tolerance on the profile gets looser. Under EN 755-9 the dimensional tolerance is indexed to the circumscribing circle — the diameter of the circle that just contains the section — not to the side length.

A 20 × 20 is held to ±0.30 mm and a 100 × 100 to ±0.90 mm — three times looser. For one member that is irrelevant. For a frame assembled from twenty of them, with brackets that have their own tolerances, it accumulates, and it is one of the reasons a large frame needs a squaring step at assembly that a small one does not.

Frequently asked questions