Most of the embodied carbon in a sheet-metal part comes from the metal itself, and the amount of metal is set mostly by one number on the drawing: the thickness. When a panel feels flimsy, the usual reaction is to go up a gauge. It works, but it is the most expensive way to buy stiffness, in money, in weight and in carbon. A flange, a return or a pressed swage will usually do the same job with less material.
This article explains why, works through a simple example, and shows how to put a carbon figure on the difference without overstating it.
Mass is linear, stiffness is cubic
The mass of a flat sheet is proportional to its thickness. Its bending stiffness is not. For a plate of width b and thickness t, the second moment of area, which governs how much it deflects, is:
I = b × t³ ÷ 12
Mass per metre = b × t × ρρ is the density, about 7850 kg/m³ for carbon steel. For a wide plate the bending stiffness per unit width is D = E × t³ ÷ (12 × (1 − ν²)), which also scales with t³.
Doubling the thickness doubles the mass but makes a flat plate eight times stiffer. That sounds like an argument for thicker sheet, until you notice that stiffness grows with the square of distance from the neutral axis. Moving a small amount of material a long way from the neutral axis, by folding it into a flange, achieves far more than spreading extra thickness evenly across the whole plate.
A worked example: 3 mm flat against 2 mm flanged
Take a 300 mm wide steel shelf or cover plate that spans between two supports and bends about its horizontal axis. Compare a 3 mm flat plate with a 2 mm plate that has a 25 mm flange folded down along each long edge, keeping the same 300 mm overall width. Bend radii are ignored to keep the arithmetic clear; including them changes the results by only a few percent.
Section A: 3 mm flat plate
- Area = 300 × 3 = 900 mm²
- I = 300 × 3³ ÷ 12 = 675 mm⁴
- Mass = 900 mm² × 7850 kg/m³ = 7.07 kg per metre of span
Section B: 2 mm with two 25 mm flanges
Split the section into the top web (300 × 2 mm) and two flanges (2 × 23 mm each, below the web). Measure heights from the flange tips.
- Web: area 600 mm², centroid at 24.0 mm. Flanges: area 2 × 46 = 92 mm², centroid at 11.5 mm. Total area = 692 mm².
- Neutral axis: (600 × 24.0 + 92 × 11.5) ÷ 692 = 15 458 ÷ 692 = 22.34 mm above the flange tips.
- Web: I = 300 × 2³ ÷ 12 + 600 × (24.0 − 22.34)² = 200 + 1 657 = 1 857 mm⁴
- Flanges: I = 2 × (2 × 23³ ÷ 12) + 92 × (22.34 − 11.5)² = 4 056 + 10 806 = 14 862 mm⁴
- Total I = 1 857 + 14 862 ≈ 16 720 mm⁴
- Mass = 692 mm² × 7850 kg/m³ = 5.43 kg per metre of span
| Section | Area (mm²) | I (mm⁴) | Mass (kg/m) | Stiffness vs A | Mass vs A |
|---|---|---|---|---|---|
| 2 mm flat | 600 | 200 | 4.71 | 0.30× | 0.67× |
| A: 3 mm flat | 900 | 675 | 7.07 | 1.0× | 1.0× |
| 4 mm flat | 1 200 | 1 600 | 9.42 | 2.4× | 1.33× |
| B: 2 mm, 25 mm flanges | 692 | 16 720 | 5.43 | 24.8× | 0.77× |
| 1.5 mm, 25 mm flanges | 521 | 12 850 | 4.09 | 19.0× | 0.58× |
The flanged 2 mm section uses 23% less steel than the 3 mm flat plate and is roughly 25 times stiffer. To reach the same stiffness with a flat plate you would need about 8.7 mm, nearly three times the mass. Even 1.5 mm with flanges is far stiffer than 3 mm flat, although at that thickness you should check that the flanges and web do not buckle locally.
Stiffness is not the only check. Confirm bending stress using the section modulus, remember that the free edges of the flanges go into compression if the load reverses, and for thin-gauge steel members use the effective-width approach in EN 1993-1-3 where local buckling could govern. The point of the example is the order of magnitude: geometry beats thickness by a wide margin.
Stiffening features that cost very little
- Flanges: the biggest gain for the least effort. A press brake adds one in seconds, and the extra blank width is small.
- Returns and lips: a short return on the end of a flange stiffens its free edge, which helps against local buckling and makes the edge safer to handle.
- Hems: a folded-back edge stiffens and deburrs an edge at the same time, at the cost of a little extra material and a two-stage fold.
- Swages and beads: pressed ribs stiffen the middle of a large panel where a flange cannot reach. They need tooling, so they suit batch work better than one-offs.
- Return to the supports: often a flimsy panel is really an unsupported span. An extra fixing or a fold that bears on the frame can halve the span, and deflection falls with the cube or fourth power of span.
Turning mass into embodied carbon
The first-order estimate of embodied carbon is simple:
Embodied carbon (kg CO₂e) = mass (kg) × emission factor (kg CO₂e per kg)Use the cradle-to-gate (A1 to A3) figure from the supplier’s Environmental Product Declaration, published to EN 15804, wherever one is available.
The emission factor is where care is needed. Published figures for steel vary widely with the production route and the source. Primary steel from the blast furnace and basic oxygen route (BF-BOF) is commonly reported at around 2 kg CO₂e per kg or more, while scrap-based electric arc furnace (EAF) steel can be well under 1 kg CO₂e per kg, depending heavily on the electricity supply and the scrap content. Aluminium spans an even wider range between recycled and primary metal. Treat any generic figure as indicative only, and ask your stockist or mill for the EPD that covers the material you are actually buying.
To illustrate the scale, suppose the shelf in the example is made in a batch of 500 at 1.2 m long, 600 m in total. Going from 3 mm flat to 2 mm flanged saves 1.63 kg per metre, or about 980 kg of steel. At an illustrative factor of 2 kg CO₂e per kg, that is roughly 2 tonnes CO₂e avoided, along with the material cost and the handling weight. With a lower-carbon EAF steel the saving in CO₂e is smaller, but the mass and cost savings remain.
Gauges, nesting and offcuts
Thickness optimisation only pays if the result can be bought and cut efficiently. A few practical points:
- Design to stock thicknesses. Common steel sheet thicknesses in the UK include 0.9, 1.0, 1.2, 1.5, 2.0, 2.5, 3.0, 4.0, 5.0 and 6.0 mm, but availability varies by grade and finish, so check with your stockist. An optimised 1.8 mm that has to be bought as 2.0 mm gains nothing.
- Keep the gauge count low. Every extra thickness on a project is another sheet to buy, store and nest. Standardising on two or three thicknesses often improves utilisation more than shaving a gauge on one part.
- Design for the sheet. Common sheet sizes include 2500 × 1250 mm and 3000 × 1500 mm. Blank sizes that nest well on the sheet your fabricator buys can cut scrap noticeably.
- Count the offcut. Scrap is usually recycled, but the steel was still made and bought. A flanged part has a slightly wider blank, so check that the extra width does not push the nest onto a larger sheet.
- Consider the process. Thinner sheet cuts faster on a laser and needs less press-brake tonnage, but very thin flanged parts can distort more during welding, so plan joints and stitch welds with care.
Checklist
- Is the thickness set by a real load case, or by habit and feel?
- Could a flange, return, hem or swage replace a step up in gauge?
- Have deflection, stress and local buckling all been checked for the thinner section?
- Is the chosen thickness a stock size for this grade and finish?
- Does the blank still nest efficiently on the sheet your fabricator uses?
- Is the embodied carbon based on the supplier’s EPD rather than a generic figure?
Reducing thickness is rarely about a single heroic redesign. It is a habit of asking whether geometry can do the job before reaching for more metal. If you would like a second pair of eyes on a part that feels heavier than it needs to be, send us your drawings.