
ID	0436
Class	KNOWN-LIMITATION
Severity	2
Status	blocked

Models: Cable/LumpedMass, Cable/Branched

Found: 2026-10-29, feature me-cable-docs (review 2026-10-24, MARE-N2; owner ruling D21)

Decision needed: whether the numerical damping should act on the node velocity relative to the water (v - u) instead of the absolute velocity, with the default (1 per second) kept or lowered. It changes the results of every cable that has current or waves in its input.

Evidence

src/cable/subroutines/InternalCable.cpp:560: adXDot[m_iStateVel[node] + index] = (adForceA[index] + adForceB[index]) / inertia - m_CableSpec.numericalDamping * adX[m_iStateVel[node] + index]; and the Jacobian at :300 (J[idx * nStates + idx] -= m_CableSpec.numericalDamping). adX[vel] is the absolute node velocity. The default is 1 per second (cable_defaults::kNumericalDamping, src/cable/subroutines/InternalCable.h); the tables of Cable/LumpedMass and Cable/Branched state that default since D21.

Effect

A cable that hangs in a steady current moves with the water at rest relative to it, yet the damping force -k m v is not zero: it drags the cable towards the earth frame, as a second drag with a rate of 1 per second on every node, independent of the diameter and the drag coefficients. In a current of 1 m/s the extra force on a node of the 0.01 m, 1100 kg/m3 default cable is of the order of the node's inertia times 1 m/s per second, which can exceed the hydrodynamic drag. The effect is not run here.

Possible fix

(a) Damp the velocity relative to the water: -k m (v - u). Cables in still water give the same numbers; cables in current or waves change, and the Jacobian needs u (its velocity derivative is unchanged). (b) Keep the absolute damping and say so (done in the docs by D21). (c) Lower the default, or make it 0, with the absolute damping kept for explicit-step runs that need it.

Test that would prove it

A cable that drifts with a uniform current from rest relative to the water: with (a) its node accelerations stay 0 and its end forces are the still-water ones; today the numerical damping gives a nonzero force.

Risk

Changes baselines of every cable test and example with a current or waves; the owner decides (physics). Fix and moves.md rows belong to a feature of their own.