DNV-RP-F105 first-mode free-span screening

DnvRpF105FreeSpanScreeningKernel provides a transparent preliminary screen for the current DNV-RP-F105 2025-12 basis. It calculates a simply supported Euler-Bernoulli first-mode frequency and common current/wave dimensionless groups. It deliberately does not reproduce licensed response or acceptance tables.

DNV identifies the current edition and public scope, including current, waves, combined loading, VIV, direct wave loading, ULS, and FLS, on its DNV-RP-F105 publisher page. Use the purchased standard and project design basis to establish applicability, response models, environmental cases, factors, and acceptance criteria.

Implemented calculation

For a simply supported span of length $L$, steel second moment $I$, effective modal mass per length $m_e$, Young’s modulus $E$, and effective axial force $T$ (positive tension, negative compression), the kernel evaluates

\[I = \frac{\pi}{64}\left(D_o^4 - (D_o - 2t)^4\right)\] \[f_n = \frac{1}{2\pi}\sqrt{\frac{EI(\pi/L)^4 + T(\pi/L)^2}{m_e}}.\]

The structural solution fails closed when compression produces a non-positive first-mode eigenvalue. The result also reports the simply supported Euler critical compression magnitude for review; NeqSim does not turn that value into a code acceptance check.

With hydrodynamic diameter $D_h$, normal current velocity $U_c$, wave orbital-velocity amplitude $U_w$, wave period $T_w$, and caller-controlled Strouhal number $St$, the result contains:

Quantity	Expression
Current shedding frequency	$f_s = St U_c / D_h$
Current frequency ratio	$f_s / f_n$
Current reduced velocity	$U_c / (f_n D_h)$
Wave frequency ratio	$(1/T_w) / f_n$
Wave reduced velocity	$U_w / (f_n D_h)$
Keulegan-Carpenter number	$U_w T_w / D_h$

The input keeps steel outside diameter separate from hydrodynamic diameter so coatings and marine growth can be represented without corrupting structural stiffness. Effective mass is an external project input and should include the accepted pipe, coating, contents, entrained-water, and added- mass basis.

Runnable Java example

StandardEdition edition = StandardEdition.defaultEdition(StandardType.DNV_RP_F105);
DnvRpF105FreeSpanScreeningKernel.Input input = DnvRpF105FreeSpanScreeningKernel.Input
    .builder(edition, "Pipeline")
    .spanLengthM(30.0)
    .steelOuterDiameterM(0.3239)
    .steelWallThicknessM(0.0206)
    .hydrodynamicDiameterM(0.3239)
    .youngsModulusPa(207.0e9)
    .effectiveMassPerLengthKgPerM(250.0)
    .effectiveAxialForceN(500000.0)
    .currentVelocityMPerS(0.8)
    .waveOrbitalVelocityAmplitudeMPerS(1.2)
    .wavePeriodS(10.0)
    .strouhalNumber(0.2)
    .lockInFrequencyRatioLower(0.8)
    .lockInFrequencyRatioUpper(1.2)
    .maxCurrentReducedVelocityForScreening(4.0)
    .maxWaveReducedVelocityForScreening(3.0)
    .spanGeometryVerified(true)
    .structuralModelVerified(true)
    .environmentalBasisVerified(true)
    .projectScreeningLimitsVerified(true)
    .build();

EngineeringCalculationResult<DnvRpF105FreeSpanAssessment> result =
    new DnvRpF105FreeSpanScreeningKernel().calculate(input, null);
DnvRpF105FreeSpanAssessment screen = result.getValue();

For these demonstration inputs, the natural frequency is about 1.06182 Hz, current reduced velocity is 2.32609, wave reduced velocity is 3.48914, and the project-controlled wave trigger requests detailed response assessment. These values are deterministic regression data, not an independent DNV benchmark or a safe-span decision.

The executed notebook uses the same API, sweeps span length to expose first-mode/reduced-velocity sensitivity, and shows a deliberately blocked unverified structural basis.

Fail-closed evidence boundary

Calculation is blocked unless all of the following are explicit and internally valid:

• exact unamended 2025-12 edition and Pipeline or AdiabaticPipe applicability;
• surveyed free-span and pipe geometry, including a hydrodynamic diameter at least as large as the steel diameter;
• accepted Young’s modulus, effective mass, effective axial force, and simply supported first-mode applicability;
• at least one positive normal current or wave velocity, with a positive period for wave input;
• caller-controlled Strouhal number, current frequency-ratio band, and current/wave reduced- velocity triggers; and
• external verification flags for geometry, structural model, environment, and project triggers.

The trigger values are recorded as caller evidence. Reaching or not reaching one does not mean that a span complies with DNV-RP-F105.

Not implemented

The kernel does not calculate or approve:

• soil, shoulder, boundary-condition, or interacting multi-span stiffness;
• mode shapes beyond the simply supported first mode;
• in-line or cross-flow VIV response amplitudes and stress ranges;
• direct wave-loading response or load combinations;
• ULS, FLS, fatigue damage, acceptance criteria, or safety factors;
• seabed proximity corrections, directionality, return-period selection, or environmental contour construction;
• sensor selection, monitoring interpretation, mitigation, buckle arrest, or intervention; or
• conformity, certification, or accountable engineering approval.

Use DnvRpC203FatigueDesignKernel only after a controlled workflow has derived applicable stress ranges and selected verified project S-N data. The F105 screen does not generate those inputs.

Legacy NeqSim span estimate

PipeMechanicalDesignCalculator.calculateAllowableSpanLength(...) remains available for backward compatibility. It uses a fixed simplified VIV-avoidance assumption and arbitrary fallback/cap behavior. It is not edition-aware and must not be reported as a DNV-RP-F105 calculation. New explicit F105 studies should use the typed kernel and retain the detailed external assessment.