How the numbers work.

The tool estimates upfront embodied carbon (modules A1–A5) for the primary timber frame of an office building: CLT floor slabs, glulam primary beams and glulam columns, plus a stated allowance for connections. It reports in kg CO₂e and in kg CO₂e per m² of gross internal area (GIA).

Module names follow EN 15978 and the RICS whole life carbon standard. Stored biogenic carbon is never deducted from the headline. It is shown separately as a ledger, as SCORS and RICS require, because the carbon is released again at end of life (module C3).

It is a concept-stage companion to the Rhino/Grasshopper definition. It is not a structural design, a RICS-compliant assessment or a SCORS rating.

Grid. In each direction the number of bays is the outline’s extent ÷ target spacing, rounded up. Bays are then made equal across the bounding box, so a 30 m length with a 7.5 m target gives four 7.5 m bays, and a 4.5 m target gives seven bays of 4.29 m.

Nodes. A grid node becomes a column when it lies inside or on the outline (a point-in-polygon test). For irregular outlines this is a counting rule, not proof of a buildable column position or load path.

Beams. Primary beams run along X. A bay is counted only where both of its columns, and the beam’s midpoint, lie inside the outline. Total beam length = sum of counted bays × storeys.

Columns. Each storey’s column runs from the slab below to the underside of the beam above: storey height − slab thickness − beam depth.

Levels. Floor levels are cumulative storey heights. The optional double-height lobby replaces the first storey height; it does not add a floor.

Carbon depends on member size, so the tool sizes each beam and column with simple checks to EN 1995-1-1 using UK National Annex values. Material properties for GL24h and GL28c come from the HASSLACHER Glued Laminated Timber technical manual (EN 14080).

Beams are single span, simply supported, and carry a floor strip equal to the Y bay. Auto searches widths of 140–280 mm and depths of 200–1,280 mm in 40 mm steps, keeping depth ≤ 4.5 × width. It picks the smallest cross-section that passes bending, shear and final deflection.

Columns are pinned at each floor and carry the interior tributary area (X bay × Y bay) from every floor above. The roof is treated as a typical floor, which is conservative. Auto picks the smallest section that passes flexural buckling about its weaker axis: square sections up to 280 mm, then 280 mm wide rectangles, then block-glued squares up to 560 mm.

Where the chosen beam also appears in HASSLACHER’s GL24h pre-design table (sections up to 360 mm deep), the tool shows the tabulated span as a cross-check.

Every factor is per m³ of product and comes from a named, published declaration. A factor that isn’t sourced is labelled as an assumption.

SCORS is the IStructE’s Structural Carbon Rating Scheme. It rates a structure’s A1–A5 carbon per m² of GIA, excluding sequestration, in 50 kg bands. An A rating is 100–150 kg CO₂e/m².

SCORS covers the whole primary structure, including substructure. This tool covers only the superstructure frame, so its position is an optimistic lower bound. A real SCORS rating for the same building will be worse once foundations, the core, stairs and full A5 are added.

LETI and RIBA 2030 targets cover the whole building, so they are not comparable with this frame-only figure.

A 30 × 24 m rectangle, four storeys at 3.6 m, 160 mm CLT, 7.5 × 6 m target grid, GL28c, default loads, 10% connections. These figures were recalculated independently from the method above and match the tool exactly.

Structure of the model. Nine aspects (substructure, superstructure, upper floors, roof, external walls, windows, internal walls, internal finishes and services) follow the structure of FCBS CARBON by FCBStudios, credited with thanks. FCBS CARBON is CC BY-NC-ND 4.0, so none of its figures are reproduced here. Every value comes from the sources below.

Material data. Generic and industry-average EPD datasets come from ÖKOBAUDAT, the German government database, which is free to use and share with the source named. The CLT and glulam figures come from the product EPDs used in version 2. The ICE database is not used: its current terms do not allow use inside a tool without a paid licence.

Modules. A1–A3 from each dataset, with stored biogenic carbon split out. A4, A5, C2, C3, C4 and D come from the dataset where declared. Otherwise A4 uses IStructE (2020) Table 4 by origin (local 0.005, national 0.032, European 0.160 kg CO₂e/kg). C2 uses 0.005 kg/kg and C3–C4 0.013 kg/kg. A5 waste = waste factor × (A1–A3 + A4 + C2–C4), using IStructE (2020) Table 5.

Building level. A5 site activity is 1,400 kg CO₂e per £100k of construction cost and C1 demolition is 3.4 kg CO₂e/m² GIA (RICS 2017). The study period is 60 years. Replacements (B4) use RICS (2017) Table 9 lifespans: roof coverings 30, cladding 30, curtain wall 35, windows 30, partitions 30, raised floors 30, floor finishes 10, ceilings 20 and services 20 years.

Operational energy (B6). GIA × energy use intensity × carbon factor for each year. Electricity uses about 0.131 kg CO₂e/kWh, from the DESNZ 2026 conversion factor update (2025 factor ≈0.177, cut 26%). Gas uses about 0.183. The decarbonising option is illustrative: a straight-line fall to 0.02 by 2035. Check these factors against the published DESNZ spreadsheet before relying on them.

Assumptions you can see and change. Foundation sizes, rebar rates, layer quantities per m² for each façade and partition build-up, the steel frame weight and the services allowance (a percentage of materials A1–A3) are tool assumptions, not published data. The inspector and the Excel export label each one.

Benchmarks. GLA offices: upfront below 950 kg CO₂e/m² (aspirational 600), A–C below 1,400 (aspirational 970), B–C below 450 (aspirational 370). UK NZCBS pilot office upfront limit: 550 by 2030. RIBA 2030 Climate Challenge (2019, non-domestic): embodied A–C below 500 and operational energy below 55 kWh/m²/yr by 2030.

Every share link records the tool version. If a link was made with an earlier version, the tool says so, because the figures may have changed.

Review status: no independent engineer has reviewed this method yet. Comments and corrections are welcome.