Crane Ground Bearing Calculator - methodology & sources

In one screen the tool answers: what load does each outrigger actually carry at this slew angle; what pressure does that put on the ground under the pad; is the pad big enough, and is the pad itself strong enough; where is the machine closest to tipping; and what is the worst slew angle - found by sweep, never assumed. Every capacity it uses is either a value you declare (with its provenance printed) or a route traced to a held document; every factor between the hoisted load and the reported pressure sits in a visible ledger.

The regulatory anchor is public US regulation. OSHA 29 CFR 1926.1402(b) requires that equipment “must not be assembled or used unless ground conditions are firm, drained, and graded to a sufficient extent so that, in conjunction (if necessary) with the use of supporting materials, the equipment manufacturer's specifications for adequate support and degree of level of the equipment are met. The requirement for the ground to be drained does not apply to marshes/wetlands.” That duty - the manufacturer's specifications, met with supporting materials where needed - is exactly the calculation this tool documents. The quotation is reproduced verbatim, second sentence included, because it is public regulation; it imposes a duty, not a formula, and is cited as documentation only.

The counts above are read from the engine's own check register at this render. The validation cases are hand-worked fixtures - the design-basis statics example, an asymmetric partial-extension case, a crawler in partial contact, a steel pad on both code routes, and three classical bearing-capacity compositions - re-run against the engine on every build.

The statics register

Route-independent rigid-body statics runs on every configuration. These checks cite public-domain mechanics only and carry no code coefficients.

Path A - declared (default, recommended)

You enter the outrigger reactions from the manufacturer's chart or lift-planning software, as a table of reaction sets keyed by slew angle - matching how planning software exports - with a single set as the one-row case. Each set is validated independently against the declared gross load:

A set that fails is reported with its discrepancy and excluded from every governing scan - the mismatch is never rescaled away, and no pressure downstream derives from a set that does not equilibrate. A declared reaction in tension is flagged, not clamped: a chart cannot output a support that pulls down.

Path B - estimated

With no chart to hand, the tool solves a rigid body on compression-only elastic supports: carrier, superstructure, counterweight, boom and hoisted load each at their plan position, the rotating components taken through the slew angle. A support that would go into tension is released and the system re-solved until every active reaction is compressive; if fewer than three supports can carry the load, the answer is a refusal, not a number. Every number derived this way is tagged ESTIMATE in the panel, the figure legend and the report, and the report's basis block names the path. Crawler machines always use this model - per-track chart reactions do not exist as an export format, and selecting crawler with a declared table raises a loud warning that the table was not used.

The worst angle is found, not assumed

On Path B the engine sweeps the full slew circle at a 1° grid with iterative refinement around every candidate extremum, and the reported worst angle is always an evaluated sample. On the default machine the peak reaction occurs with the boom at 230° - over the rear-right quarter - at 669 kN, which is 29% above the worst support at the declared over-front angle (519 kN). A calculation checked only at the angle someone guessed would miss that entirely. On Path A the same scan runs over every equilibrating declared set.

The crane's outrigger float has a small area $A_f$. Its pressure $R/A_f$ acts on the pad and governs the pad's own structure - contact, cantilever bending and shear. The pad then distributes that load to the ground over an effective area built from the float dimensions, grown through the pad thickness at the declared spread angle and capped at the physical pad edge:

ψ = 0° is the default, and it is deliberate

No spread credit is taken unless you declare an angle. At the default the effective area equals the float area and the mat's plan area earns nothing - which is why the calculator's raw default opens on a failing ground check: 1400 kPa on the ground against a 250 kPa allowable, utilisation 5.6. A 45° default would silently halve the reported pressure, and hidden generosity is precisely what this tool exists to expose. Declaring ψ is a real trade: it lowers the ground pressure and loads the pad's own bending and shear, because the claimed spread cantilever is what carries it.

The pad structural checks

The claimed overhang works as a cantilever strip from the float edge under the ground pressure: elastic strip bending ${\sigma }_b=6M/t^2$ with $M=q{s}^2/2$, peak parabolic one-way shear $\tau =1.5V/t$, and punching on the plain float perimeter - each against a user-declared allowable for timber and composite mats. A steel pad additionally gets two sourced code routes, side by side and cross-referenced:

• AISC 360-22 - strip plastic moment $M_n=F_y{t}^2/4$ (the rectangular-bar provision, whose 1.5 $F_{y}S$ cap binds exactly at equality for a unit-width strip), in allowable-strength form with ${\Omega }_b=1.67$ - chart reactions are service-level loads, so the load-factored form would mix regimes.
• EN 1993-1-1:2005 - a superseded edition, labelled - the same plastic resistance $M_{pl,Rd}=f_y{t}^2/4÷{\gamma }_{M0}$ with the recommended ${\gamma }_{M0}=1.00$ (a National Annex may alter it). The 2022 second generation is not held, so the 2005 basis is named on the card, in the report and in the sources. Because EN resistances oppose factored actions and a chart reaction is a characteristic total that cannot be decomposed into permanent and variable shares, the whole strip moment takes ${\gamma }_F=1.50$ - the higher of EN 1990's recommended action factors (from the held 2002+A1 edition, itself superseded and labelled so), the conservative envelope of the recommended set whenever both shares act unfavourably on the reaction, which is the normal case for a bearing reaction. A configuration whose self-weight relieves the governing support (${\gamma }_{G,inf}=1.00$) sits outside that envelope and needs a decomposed EN 1990 combination - the card says so. The factor is shown as its own ledger row.

The coded timber-mat route runs EN 1995-1-1:2004+A1:2008 bending, shear and compression-perpendicular-to-grain checks on user-declared characteristic strengths ($f_{m,k}$, $f_{v,k}$ and $f_{c,90,k}$ from the supplier - EN 338 / EN 14080 / EN 14374 are NOT HELD). Service class and load-duration selectors default to class 3 and short-term for an outdoor crane mat. Switch to the coded route in the inputs; until then declared allowables carry the verdict.

Route R1 - declared allowable (default)

The governing ground pressure is checked against the allowable bearing pressure from your geotechnical report, divided by your declared factor of safety. The tool applies the value; it does not derive it, and the card says so. Guidance on choosing the factor exists - industry guidance describes a factor of 1.5 to 3.0 on outrigger loading, scaled to how well the hardstand bearing capacity has been evidenced - but the source document (CIRIA C703) is not held, so that band appears beside the input as unimplemented guidance text and is never computed for you.

Route R3 - classical theory (opt-in, named)

For early screening with declared soil parameters, the classical route computes the ultimate bearing resistance from a formulation you pick by name - every factor closed-form, nothing tabulated:

• Vesić (1973), drained - Reissner/Prandtl $N_q$ and $N_c$ with $N_{\gamma }=2(N_q+1)\tan \phi$, De Beer shape factors, Hansen/Vesić depth factors.
• Meyerhof (1963), drained - the same $N_q$ and $N_c$ with $N_{\gamma }=(N_q-1)\tan (1.4\phi )$ and his $K_p$-based shape and depth factors.
• Skempton (1951), undrained - $q_{ult}=c_u{N}_c+q_0$ with the depth term capped at its published limits; total-stress, so groundwater does not enter.

On the worked example switched to this route - dry sand, φ′ = 30°, γ = 18 kN/m³, founding depth 0.5 m - the engine computes $N_q$ = 18.40, $N_c$ = 30.14, $N_{\gamma }$ = 22.40 at this render - the values the published factor tables print for that angle, which is the point: the implementation is benchmark-derived, locked to published factor tables and hand-worked compositions on every build. The ultimate resistance composes to 419 kPa; divided by the declared FoS 1.50 it gives a capacity of 279 kPa against the 214 kPa demand - utilisation 0.77.

Groundwater is handled by the effective-stress treatment - effective surcharge above the base, buoyant unit weight within the influence depth below it, the standard linear interpolation between - and the groundwater depth defaults to the surface: the conservative default, the same doctrine as ψ = 0°. Declare the real depth to relax it.

Route R2 - EN 1997-1 Annex D (informative)

BS EN 1997-1:2004 incl. corrigendum Feb 2009 is held. Route R2 implements Annex D's sample analytical bearing resistance (drained eq D.2 and undrained eq D.1) with Annex A recommended partial factors - each shown on the check card and labelled as an EN recommendation because the National Annex is NOT HELD. They are not user-overridable: where your NA sets different Nationally Determined Parameters, this route does not represent them. Annex D provides shape, inclination and base factors only - no depth factors (unlike route R3). The verdict is $V_d\le R_d$ per §6.5.2.1; demand uses the same ${\gamma }_F=1.50$ / $1.30$ envelope as the EN steel-pad route because chart reactions cannot be decomposed into permanent and variable shares. A relieving self-weight share (${\gamma }_{G,inf}=1.00$) sits outside that envelope.

All three flags use one named public-domain construction: the 45° (1H:1V) load-dispersion wedge from the edge of the bearing area. A declared excavation of depth H flags any bearing area closer than H to the face - the classic stay-back-the-depth site rule. A declared slope of height H at face angle β flags bearing areas within $H(1-1/\tan \beta )$ of the crest. A declared buried structure with its top at depth d flags bearing areas within d of its line, and anything directly above it always flags.

The verdict semantics are deliberately conservative about what a flag may claim. A passing flag is a geometric statement about the declared condition only - it carries no utilisation, so it can never displace a structural check in the governing scan. A fired flag fails the run with the reach-to-clearance ratio; a bearing area at or past the declared line fails with no ratio at all, because the words carry it. And two honest edges are stated on the cards themselves:

• A slope face at or under 45° can never fire the geometric flag for a bearing area clear of the crest - a 45° dispersion line cannot daylight on a gentler face, though a bearing area at or past the crest line itself still fails whatever the angle. That is a statement about dispersion geometry, not about the slope being safe: slope stability is out of scope by design and is not assessed.
• The buried-structure flag does not compute the pressure reaching the structure or compare it against a rated capacity - that needs a stress-attenuation analysis and the owner's data. It states, geometrically, whether the dispersion zone reaches the declared depth and line; the asset owner's requirements and a crossing/protection design govern.

Nothing in the engine multiplies a demand or divides a capacity outside this ledger. The two-level pad card spells its capacity as allowable ÷ FoS with both terms visible, and the classical route spells its own as ultimate ÷ FoS - the separation is the point: you can always see whether a margin lives in the allowable, the factor, or both.

Every gate states its reason and its remedy. The full set:

The crane manufacturer's outrigger load chart and rated-capacity chart are authoritative. This tool does not replace them; its statics are a cross-check and a pad/ground workup, never an operating margin.

Estimated reactions (Path B) assume equal outrigger stiffness unless a per-support relative stiffness is declared. Real stiffness varies with beam extension; the equal default is an assumption, not a fact.

The machine is treated as a rigid body, level and static. Dynamics enter only through the declared dynamic allowance; out-of-level enters only through the declared allowance; wind is not modelled.

Contact pressures are uniform over their bearing areas (float on pad, effective area on ground), with the float centred on the pad. Local pressure peaks from pad flexibility are not modelled.

Load spread through the pad is credited ONLY at the declared angle ψ (default 0 - no credit), capped at the physical pad edge. The declaration and its justification are the user's.

Crawler lateral distribution follows the rigid body on two elastic line supports: per-track load P/2 ± P·ȳ/s with the longitudinal moment split equally between tracks; each track is then treated independently after the split. The tipping polygon uses the track CENTRELINES, not the outer edges of the shoes - the conventional and conservative fulcrum. Partial contact is reported only while at least L/6 of the track still bears: the triangular peak 2·P_t/(b·L′) diverges as the contact closes, so beyond that the state is INDETERMINATE rather than an ever-larger number.

Every capacity on the declared routes - allowable ground bearing pressure, pad contact/bending/shear allowables - is a user-declared value (geotechnical report, manufacturer rating). Its provenance and validity are the user's responsibility; this tool applies it, it does not derive it.

Route R3 soil parameters (φ′, c′/c_u, γ, embedment, groundwater) are user-declared values from the site's geotechnical information. The computed ultimate resistance is a named classical formulation, not a code check, and a site-specific geotechnical investigation SUPERSEDES it.

Proximity checks are geometric zone-of-influence FLAGS built from the declared geometry only (the 45° load-dispersion construction). They do not analyse the excavation face, its support, surcharge on a retaining structure, slope stability, or the buried structure's capacity - a fired flag means a temporary works design is required.

Outside scope by design: settlement (bearing capacity is not serviceability), dynamic and impact effects beyond the declared allowance, wind on the crane and load, slope stability and global site stability, and any crane machine database.

Out of scope, with reasons

• Settlement. Bearing capacity is not serviceability: a pad can pass every check here and still settle more than the lift can tolerate. Settlement needs ground stiffness data this tool does not ask for.
• Wind on the crane and load. Getting the machine's exposed area right is a machine-specific question that pushes toward the crane database this tool permanently excludes. The chart's wind limits govern operations.
• Slope stability and global site stability. A different discipline. The proximity flag tells you when a temporary works design is required; it never substitutes for one.
• Dynamics beyond the declared allowance. Real dynamic effects are machine- and duty-dependent; the declared factor is a visible ledger row, never a hidden default.
• Any crane machine database. Permanent exclusion, not a roadmap item - the chart in the cab is authoritative, and a transcription would invite misplaced trust.