StruCla against SCIA Engineer: steel frame on plinth walls and a foundation slab
A steel support frame (11 HEB140 members, S235) stands on three concrete plinth walls (0.9 m high, 0.4 m thick) that stand on a foundation slab of 2.0 × 3.4 × 0.4 m. All concrete is C30/37. The slab lies on gravel subsoil (bedding 20 MN/m³ vertical and 20 MN/m³ horizontal). This is a limited comparison: the SCIA output holds the results of the load case self weight (LC1) only, so only LC1 is compared. The wind, friction and pipe loads of the other four load cases are in the SCIA input but have no SCIA results here, so the horizontal loading of the plinth is not validated by this page. The plate and wall results come from StruCla's new shell solver (plates in different planes joined along shared edges); it is not yet part of the online app. Project name, client, plant, licence data and author have been removed from the SCIA output.
Green: equal within the larger of 0.15 and 20 % of the SCIA value (kN, kNm; the steel frame: within the larger of 0.02 and 5 %). Red: a larger difference; it is described below. Each cell shows SCIA on top and StruCla below.
1. Model and input
| SCIA Engineer | StruCla | |
|---|---|---|
| Steel frame | 11 members HEB140, S235, 32 nodes; the loads on the members are given in the SCIA output (sheet 3) | the same; in this comparison the frame carries only its self weight (LC1) and its foot reactions act on the concrete |
| Plinth walls | S1, S2 (1.3 m long) and S3 (1.25 m), 0.9 m high, 0.4 m thick, C30/37 (E = 32.8 GPa, ν = 0.2, 2500 kg/m³) | the same; E = 33 GPa, 25 kN/m³ (SCIA: 24.525 kN/m³) |
| Foundation slab | S4, 2.0 × 3.4 m, 0.4 m thick, C30/37, the walls stand on it along the internal edges | the same; shared edges with the walls are rigid |
| Subsoil | surface support: gravel/slightly silty/loose (NEN 6740), C1x = C1y = C1z = 20 MN/m³ under the slab | spring bedding under S4: 20 000 kN/m³ normal and 20 000 kN/m³ tangential |
| Elements | 2D elements of about 0.5 m | S4: 0.125 m; S1: the same 6 elements as SCIA (3 columns of 0.25 / 0.80 / 0.25 m, 2 rows of 0.45 m) |
| Solver | SCIA Engineer 21.1 | StruCla shell solver: membrane Q4 with incompatible modes and Mindlin plate bending (MITC4), 6 degrees of freedom per node |
The shell solver was first checked against closed-form solutions (simply supported and clamped square plate, deep beam, a folded plate in equilibrium); all within 1 %.
2. Steel frame, forces for self weight (SCIA sheet 4), kN and kNm
SCIA prints N, Vy, Vz, Mx, My and Mz at eight extreme sections. The table gives N, the larger shear force and the larger bending moment per section (the local axes of the two programs differ). SCIA takes 7850 kg/m³ times 9.81 m/s², which is 77.0 kN/m³; StruCla uses 78.5 kN/m³ for steel. As the only load is self weight, the StruCla values below are multiplied by 77.0 / 78.5, which is exact.
| Member | dx (m) | N | V | M |
|---|---|---|---|---|
| B1 | 0.000 | −1.18 −1.18 | 0.01 0.03 | 0.01 0.00 |
| B4 | 0.000 | 0.01 0.03 | 0.18 0.19 | 0.04 0.05 |
| B15 | 1.000 | −0.01 −0.01 | 0.17 0.17 | 0.01 0.01 |
| B6 | 0.000 | 0.00 0.00 | 0.26 0.26 | 0.06 0.06 |
| B1 | 0.600+ | −0.79 −0.79 | 0.01 0.00 | 0.07 0.07 |
| B10 | 0.500 | 0.00 0.01 | 0.00 0.00 | 0.04 0.04 |
| B2 | 1.700 | −0.17 −0.17 | 0.01 0.01 | 0.01 0.01 |
| B1 | 1.700 | −0.17 −0.17 | 0.01 0.01 | 0.01 0.01 |
The base of the columns is fixed on the plinth in this calculation. The displacements of SCIA (0.7 to 0.9 mm, sheet 4) are those of the whole structure on the subsoil; StruCla's shell solver settles the structure by 0.79 to 0.91 mm.
3. Foundation slab S4, in the element centres (SCIA sheet "2D internal forces", Basic magnitudes, LC1)
For each of the 12 elements of SCIA's output the nearest element of StruCla is taken (the distance between the element centres is at most 0.18 m). Moments in kNm/m, shear forces and membrane forces in kN/m, in the local axes of the slab; the sign of the shear force of SCIA is opposite to that of StruCla and has been reversed.
| Element | x (m) | y (m) | mx | my | mxy | vx | vy | nx | ny | nxy |
|---|---|---|---|---|---|---|---|---|---|---|
| E50 | 5.117 | -2.420 | −0.14 −0.07 | −1.65 −1.62 | 0.43 0.42 | 0.48 0.75 | −0.89 −0.95 | 0.02 0.02 | 0.11 0.15 | −0.01 −0.03 |
| E48 | 5.661 | -1.918 | 0.13 0.26 | −1.89 −1.94 | 0.48 0.51 | 0.23 0.17 | 0.11 −0.30 | 0.08 0.02 | 0.31 0.20 | −0.07 0.01 |
| E19 | 4.706 | -3.761 | 0.22 0.12 | 0.15 0.24 | −0.13 −0.11 | 1.11 1.45 | 0.96 0.87 | 0.09 −0.01 | 0.16 0.18 | −0.41 −0.46 |
| E33 | 6.188 | -2.846 | 0.09 0.50 | −0.22 −0.26 | 0.17 0.17 | −3.42 −3.03 | −1.98 −1.72 | 0.09 0.03 | −0.12 −0.01 | −0.15 −0.14 |
| E49 | 5.608 | -2.910 | 0.68 0.47 | −0.34 −0.41 | 0.14 0.08 | 3.38 2.73 | −1.49 −1.66 | 0.77 0.69 | 0.14 0.27 | 0.24 0.09 |
| E42 | 5.093 | -3.314 | 0.32 0.41 | 0.05 0.13 | 0.00 −0.06 | 0.49 0.25 | −3.50 −3.83 | 1.36 1.13 | 0.01 0.01 | 0.67 0.71 |
| E46 | 5.091 | -1.283 | 0.45 0.39 | −0.35 −0.68 | 0.28 0.34 | −0.29 −0.32 | 4.01 3.57 | 1.15 0.66 | −0.05 −0.01 | −0.37 −0.23 |
| E34 | 6.375 | -3.358 | 0.29 0.27 | 0.51 0.41 | 0.10 0.05 | −2.12 −2.51 | −1.72 −1.43 | −0.39 −0.05 | 0.35 0.21 | −0.44 −0.41 |
| E38 | 5.575 | -3.688 | 0.52 0.68 | 0.45 0.37 | −0.08 −0.13 | 0.56 0.48 | 1.51 2.08 | 2.15 1.72 | −0.47 −0.23 | 0.04 0.00 |
| E32 | 6.188 | -2.395 | 0.47 0.32 | −1.10 −1.35 | 0.11 0.16 | −1.82 −1.78 | −2.91 −2.56 | −0.03 0.06 | −0.50 −0.15 | −0.05 −0.08 |
| E40 | 5.958 | -3.392 | 1.09 1.56 | 0.40 0.57 | 0.10 0.09 | −1.37 −3.37 | −1.17 −1.20 | 1.99 0.39 | 0.95 0.68 | 0.00 −0.40 |
| E35 | 6.167 | -3.225 | 0.80 0.61 | 0.39 0.26 | 0.06 0.10 | −2.11 −2.96 | −1.16 −1.48 | 0.44 0.20 | 0.12 0.12 | −0.53 −0.26 |
71 of 96 values equal (74 %). The moments and the membrane forces away from the plinth agree well (my of element 50: −1.65 against −1.62; of element 48: −1.89 against −1.94). The differences concentrate in the elements near the east end of the plinth walls (x = 6.0 to 6.4 m), for example elements 33, 40 and 35.
4. Plinth wall S1, in the element centres (LC1)
StruCla's mesh has the same 6 elements as SCIA. The centres lie up to 0.1 m apart; in the wall the forces change quickly near the base, which accounts for part of the differences. The wall lies in the plane of the first edge: nx is horizontal, ny vertical. The sign of nxy is that of SCIA; for vx and vy the absolute values are compared (the axes of the wall differ between the programs).
| Element | x (m) | z (m) | mx | my | mxy | vx | vy | nx | ny | nxy |
|---|---|---|---|---|---|---|---|---|---|---|
| E5 | 6.000 | 0.675 | −0.02 0.02 | −0.06 0.02 | −0.03 −0.04 | 0.01 0.07 | 0.08 −0.17 | 0.28 −0.25 | −3.80 −3.41 | −0.53 −0.60 |
| E1 | 4.984 | 0.675 | 0.00 0.00 | 0.01 0.00 | 0.00 −0.01 | −0.03 −0.01 | −0.01 0.05 | −0.02 −0.16 | −3.30 −2.68 | 0.69 0.39 |
| E4 | 5.578 | 0.225 | 0.03 −0.04 | −0.04 0.04 | −0.07 −0.06 | 0.28 0.21 | −0.40 0.32 | −1.43 −1.21 | −4.13 −4.93 | 0.51 1.06 |
| E2 | 5.509 | 0.675 | 0.09 −0.11 | −0.02 −0.01 | −0.04 −0.04 | 0.10 0.11 | −0.04 −0.01 | −2.50 −2.14 | −2.68 −3.28 | 0.42 0.65 |
| E6 | 6.000 | 0.225 | 0.04 0.00 | −0.07 0.07 | −0.01 −0.03 | 0.17 0.07 | −0.05 −0.08 | −3.12 −1.52 | −12.31 −12.81 | −3.95 −3.78 |
| E3 | 5.053 | 0.225 | 0.02 0.00 | 0.01 −0.03 | −0.01 −0.02 | 0.17 0.02 | −0.05 −0.02 | −2.54 −1.69 | −9.42 −12.53 | 3.07 3.74 |
38 of 48 values equal (79 %). The vertical force at the base is equal: −12.3 kN/m (SCIA) against −12.8 kN/m (StruCla) in the east element, −9.4 against −12.5 in the west element (the element centre of StruCla is 0.1 m closer to the corner).
5. What differs, and why
- Horizontal force at the base of the wall S1. SCIA has nx = −3.12 and −2.54 kN/m in the two lower elements at the ends, StruCla −1.52 and −1.69. The ratio nx/ny is 0.25 in SCIA and 0.12 in StruCla. We varied the slab and wall thickness (0.2 to 0.8 m), the stiffness of the subsoil (2 000 to 2 000 000 kN/m³), the mesh and the elements, without effect; only the Poisson ratio changes it (0.2: 0.12, 0.4: 0.20), and SCIA's concrete has ν = 0.2. This is not explained. The difference is about 1.5 kN/m on a wall of 0.4 m, so 0.004 N/mm².
- Values at the nodes. At the corner where the plinth wall meets the slab the force is a point singularity and depends on the mesh: the vertical force at the base corner of S1 is −14.9, −22.1 and −30.7 kN/m for elements of 0.25, 0.125 and 0.0625 m in StruCla; SCIA's nodal value is −28.1. These nodal values (and the shear forces next to the wall lines, where SCIA reports one value per element) are therefore not compared; the element centres are.
- East end of the plinth walls. The elements of S4 at x = 6.0 to 6.4 m differ most (for example element 40: nx 1.99 in SCIA, 0.42 in StruCla). It is probably related to the first point; the cause is not known.
- Material and weights. StruCla takes E = 33 GPa for C30/37 (SCIA 32.8 GPa) and 25 kN/m³ for concrete (SCIA 24.525 kN/m³): a difference of 2 % in the weight, small against the differences above.
- Not validated here: the load cases LC2 to LC5 (pipes, friction, wind), the combinations, the reinforcement design and the contact stresses; the SCIA output has no results for them.
6. How the comparison was made
All values of the three tables are in the check sheet (also as json), with the difference per value. The SCIA values are those of the SCIA output (sheet 4 and the tables "2D internal forces" for the slab and the wall with the location "in centres", basic magnitudes, which are the same output selected for S4 and S1). StruCla's values come from its shell solver; the solver is not in the online app yet, so the model file is not offered for download.