IFC showcase: Borgund stave church, 1180

3D model

The viewer opens stave-church.glb, the glTF that NativeBIM writes from the IFC file. The IFC file is stave-church.ifc.

What is in the model

Requirements

The model meets every requirement of a timber building. Each is measured on the model’s geometry by casting rays at it from sample points 0.25 m apart.

Drawing set

Draft drawing set generated from the model: stave-church-drawings.pdf (PDF, 3 sheets).

• G-001 COVER SHEET AND SHEET INDEX
• A-101 CHURCH FLOOR PLAN
• A-601 DOOR SCHEDULE

Elevations are not generated yet, so this drawing set has plans and schedules only. Sections are not generated yet, so this drawing set has plans and schedules only.

Why it stands, and what it can take

1087 frame members and 534 joints read off the 1861 survey, solved for 13 load cases and checked in 28 Eurocode combinations. The church's own weight is 47.8 t; under 1.35 times that weight the worst member check is 0.16 (1.00 is design strength). Snow does not govern: the nave roof is 56 degrees, so Eurocode's shape coefficient there is only 0.10 (0.80 on a flat roof). Wind does: the governing check is 0.59 (Mast, 1.2 G + 1.5 W (from the north, cpi 0.2)). Pushed further, the weakest member reaches its design strength (with Eurocode's safety factors) at 33.9 m/s and its characteristic strength (without them) at 52.2 m/s; the church would slide on its stones at 30.4 m/s and tip at 43.7 m/s (basic wind velocity, terrain II).

Geometry: the mast core, nave walls, gallery, chancel, apse and turret are measured off Schirmer's 1861 survey (plan, cross-section and longitudinal section); the section sizes of the rectangular timbers, the plank thickness and the roofing weight are chosen. The gallery's solid parapet (0.94 m) and open arcade, and the side and west porches, are taken from the same sheets; as in the photographs, each porch roof runs back over the gallery roof to the nave wall, meeting it in valleys. The roofs and walls the viewer shows are the load panels themselves, so the analysis is solved on the same shapes: every roof runs on into a 0.45 m eave (chosen; the eave's load goes to the rafter feet, its small lever arm over the plate is not modelled), the turret roof, the spire and the apse tower's cap flare out at the eaves (bell-cast, with round-tipped scale shingles), the apse is round in plan with a cone roof up to a round drum on its tower, and the gallery runs round the east end on a curve through the surveyed posts. These follow the photographs; the survey gives the frame, not the curves. Drawn but not loaded: the boards and shingles' thickness outside the frame, the dragon heads, crosses, bargeboards and porch lattice (traced off Schirmer's detail sheets), and the gallery arcade, portals with their oak doors and iron hinges, belfry arcade, porch arches and the hoods at the gable feet (read off the photographs with chosen sizes).

Loads: own weight (pine 500 kg/m3), roofing 450 Pa, planks 392 Pa, floors 350 Pa; ground snow 3 kPa with Eurocode's shape coefficient on each roof plane, balanced and unbalanced, and a held-snow case (0.8 on every roof); wind 26 m/s in four directions with internal pressure +0.2 and -0.3. These are chosen values, not Borgund's code values.

Load path: each roof or wall panel is cut into 15 cm cells; each cell's force goes to the two members its boards span between (rafters, plates, girts or posts) by the lever rule, and lands on the frame as member point loads (23435 of them).

Joints: masts are continuous; every pegged beam end is a rotational spring of 100 kN.m/rad; knees, crosses and roof braces are pin-ended; the sills sit on the stones with no anchorage, so lifting, sliding and tipping are limits of the building, not of a fixing. The stave walls are bands of 80 mm planks with their out-of-plane stiffness, plus an in-plane racking stiffness of 200 kN/m per metre of wall (pin-ended equivalent diagonals in each bay, analysis only): a chosen value, about one hundredth of what the planks would give if they could not slip in their grooves. The bracing table shows the results with none and with twice as much.

Checks: EN 1995-1-1 compression or tension with bending (buckling over the length between joints), and shear, at 13 points along every member, with C24 strengths, kmod for the shortest load in the combination and gamma_M 1.3. Not checked: the joints themselves (except the mast foot, below), lateral-torsional buckling, bearing across the grain at the sills.

Mast foot joint (fea/): the worst design uplift at a mast foot is 44.4 kN (Mast 7, 1.0 G + 1.5 W from the north, cpi -0.3). One pegged tenon is checked for it by hand and with two solid CalculiX models, the tenon and the sill, solved once for 38.2 kN and committed as recorded output (provenance in fea/RECORDED.json); the solids are linear elastic and the hand checks are force over an area, so the table scales every value to the uplift the frame gives now. The joint sizes are chosen as typical for the period (310 mm mast, 80 x 200 mm tenon 200 mm deep, one 30 mm peg), not measured, and the solids are isotropic. The tenon is lightly used (its FEA averages agree with the hand values). The peg in double shear is not: 10.10 at design and 3.98 at characteristic level. The foot starts to lift at 12.4 m/s (design level) and one peg reaches its design strength at 14.3 m/s; the sill cheeks bearing on the peg across their grain follow at 14.6 m/s. Both checks are conservative (a peg breaks across its own grain, several times stronger than the shear strength used here, and Eurocode 5 dowel embedment strength is several times fc,90,k), but the reading stands: a pegged foot is not a hold-down. The frame model ties every mast foot to the ground; at Borgund the masts stand on the groundsill frame, which can only hold them down with its own weight and the walls'.

The wind limits are unfactored for lifting, sliding and tipping (the wind at which the church's own weight is used up), and at design level for the members (1.0 G or 1.2 G with 1.5 W). Equilibrium residual of the dead load solve: 8.5e-14.

A screening study of an illustrative model, not an assessment of the real building and not engineering advice. The 1861 survey drawings shown are in the public domain (credits in the figure captions).

Sources

Where each dimension comes from. Anything marked as chosen for the sample is an assumption, not a measurement.

• Dragon head (sheet 0114), porch cross, bargeboard width and porch gable lattice (north elevation, sheet 0010): the carved boards in showcases/stave-church/ornaments.js. H. M. Schirmer, measured drawings of Borgund stave church, 1861, Riksantikvaren RADT 286 sheets 0114 and 0010, public domain.
• Porch roofs running back over the gallery roof, pointed shingles, colours, gallery arcade, portals, belfry arcade and gable hood (read off, not reproduced). Photographs of Borgund stave church on Wikimedia Commons by Bjørn Erik Pedersen (2016, 2023, CC BY-SA 4.0), mendhak (2018, CC BY-SA 2.0) and Ximonic (2013, CC BY-SA 3.0).
• Plan, cross-section and longitudinal section; every coordinate of the frame (pixel readings in showcases/stave-church/geometry.js). H. M. Schirmer, measured drawings of Borgund stave church, 1861, Riksantikvaren RADT 286 sheets 0127, 0017 and 0018, public domain.
• Nave 7.35 x 6.10 m, chancel 3.4 m square, apse 1.88 x 2.98 m, nave about 11 m floor to ridge. Kringom, Borgund stave church (after Henden Aaraas, Vengen and Gjerde, 1999).
• Timber felled 1180/81 (dendrochronology); 12 masts with two half-posts; tang, crosses and scissor roof. Store norske leksikon, Borgund stavkirke; Wikipedia, Borgund Stave Church.
• Over 1200 wooden nails in the main structure; knee brackets of root wood. Fortidsminneforeningen, Digitale Borgund (2021).
• Wind: peak velocity pressure, terrain II, Tables 7.1, 7.3a/b, 7.4a/b, Figures 7.4 to 7.8; snow shape coefficient Table 5.2. EN 1991-1-4:2005+A1:2010 and EN 1991-1-3:2003 recommended values (not the Norwegian national annex).
• Combinations 6.10b and Table A1.2(A); timber C24, kmod, gamma_M 1.3, member checks 6.1 to 6.3. EN 1990:2002, EN 338:2016, EN 1995-1-1:2004 recommended values.
• Reference wind 26 m/s, ground snow 3.0 kPa, roofing 0.45 kPa, planks 80 mm, floor 0.35 kPa, joint stiffness 100 kN.m/rad, friction 0.5, rectangular section sizes. Chosen for this study, not Borgund's code values and not measured.

Built 2026-10-31 with node tools/showcase/build.js stave-church. Re-running the command gives the same bytes.

Notices