Code validation¶
Clause 10.2 of the Italian NTC 2018 asks the designer to assess the reliability of the calculation code they use. It is not a duty of the software producer: it is your judgement. Our task is to give you the material to form it.
For MP NX that material is the validation document: fields of application and limits, theoretical bases with references, fully solved and reproducible test cases. This page summarises it. It is a validation, carried out by those who develop the program and verifiable by anyone: no third-party body is involved.
Fields of application and limits¶
MP NX computes a single reinforced-concrete pile, bored or driven, and a single micropile. The functions are listed on the home page.
It does not cover: pile groups and group efficiency; screw piles; jet grouting; steel or timber piles; micropile networks, raked micropiles and critical buckling load; check of the composite tubular section; non-circular sections in the structural check; pressuremeter tests and dynamic formulae; Fleming's settlement; the full NTC seismic hazard and kinematic moments.
The Winkler model captures neither the interaction between springs nor soil yielding beyond the removal of springs: piles with large displacements need p-y or continuum analyses.
Theoretical bases¶
Berezantsev et al. (1961), Terzaghi (1943), Janbu (1976), Hansen (1970) and Vesic (1977) for Nq and Nc; Broms (1964) in Viggiani's formulation (1999) for the lateral load; Poulos and Davis (1980) for the settlement; Bowles (1996) and Chiarugi-Maia (1970) for the modulus of subgrade reaction; NTC 2018 with Circular 7/2019, EN 1997-1 and EN 1992-1-1 for the factors and the section check.
Test cases¶
The document holds 18 cases, all repeated as automated tests at every release.
| Cases | Subject | Reference | Difference |
|---|---|---|---|
| 1 | Nq and Nc at φ = 30° for all methods | literature formulae | < 1 % (< 2 % Berezantsev with D = 1 m) |
| 2–4 | pile in sand, in undrained clay, in rock | hand calculation | within ± 1 % |
| 5 | Broms, six cases (cohesive and cohesionless, free and fixed head) | closed forms | ± 1 % |
| 6 | Poulos and Davis | table of the method | 6.1 mm expected, 6.0–6.25 computed |
| 7 | NTC chain: ξ, γb, γs, tension, global factor | NTC 2018 tab. 6.4.II and 6.4.IV | exact |
| 8–10 | the three design examples | they load and calculate | — |
| 11–12 | FEM of Bowles example 16.10; Bowles Ks by hand | desktop and hand calculation | 0.0 % on the solver; 0.4 % through the archive |
| 13 | ultimate moment Ø 80 cm, 12 Ø 20 | non-dimensional design charts | 455 kNm expected and computed |
| 14–18 | five files of the MP desktop | desktop program | see below |
Wherever it exists, the reference is the closed form from the literature recomputed by hand: anyone can verify it and it does not depend on us. The comparison with the desktop is added to it, it does not replace it.
Differences, explained¶
Differences are not hidden: each has a declared cause.
- Pile weight, + 2 % (case 14). The desktop, in its kg system, uses 2 500 kg/m³ for C20/25 (24.52 kN/m³); the SI archive has 25 kN/m³. Base and shaft match; once the weight difference is removed, Q and Rd are back within 0.1 %. The difference is conservative.
- Elastic modulus, 0.4 % (case 11). Same origin: the two desktop archives have different E for the same class. With the same modulus the solver matches to the fourth digit.
- Shear resistances, − 2 % (case 16). VRcd and VRsd are lower than the desktop at all nodes. It corresponds to an effective depth smaller by one stirrup diameter: the desktop uses a compiled version of the solver, MP NX the sources. The difference is conservative.
- Cases 15, 17 and 18: 0.0 % on ultimate load and design resistances; in case 15 also on the settlement, in case 18 also on Broms.
To reproduce a desktop file in the technical system, put the values of the technical archive in the project archive (for C20/25: Ec = 29 380.7 MPa, γ = 24.52 kN/m³). The difference that remains is then due to the calculation, not to the archives.
Declared choices¶
Some behaviours of the desktop program have been kept and are listed in the document: the taper factor Fw for φ ≥ 30°, the Vesic and Sano seismic correction starting from the unreduced φ, the adhesion not removed when the founding plane falls within the first layer, Rd with the pile weight of the first vertical. A desktop error in the description of Broms' mechanism has instead been fixed; Hmax does not change.
Where to find it¶
The five validation cases can be downloaded from the app: see Example projects. The page on the validation of GeoStru codes opens from the ? button, Resources tab, item Validation of the calculation code.
Found an error on this page? Let us know or open a Pull Request.