Skip to content

fix: seabed contact for clump points and buoys; note EI = 0 folds - #28

Merged
SMI-Lab-Inha merged 5 commits into
devfrom
fix/points-seabed-folds
Oct 20, 2026
Merged

SMI-Lab-Inha merged 5 commits into
devfrom
fix/points-seabed-folds

Conversation

@SMI-Lab-Inha

Copy link
Copy Markdown
Owner

Summary

Clump weights and buoys now rest on the seabed instead of sinking into it, and a tension-only line whose static layout folds back on the seabed is named instead of passing silently.

  • Seabed contact for points. Only the end nodes of the attached lines held a Free/Connect point on the seabed, so a heavy clump sank into it: a 50 t clump between two polyester lines ended 2.8 m below the seabed, a 20 t clump on chain 1.0 m.
    • A point with Mass or Vol, and a Point3 buoy, now rests on a footprint A = V^(2/3) with the line contact law (kBot·A, cBot·A), along the surface normal on a bathymetry.
    • Applied in the point static equilibrium and in the monolithic, staggered, OpenFAST and C API point steps. The point routes still reject frictionMu, so the contact is frictionless.
    • The 50 t clump now rests 0.95 m deep, the 20 t clump 0.67 m. Of the examples only clump_weight_free_point.dat changes (clump 4.7 cm deep instead of 6.6 cm; anchor tension 186 kN to 92 kN). Decks whose points stay clear of the seabed give identical output.
    • OpenFAST linearisation of a deck with a Point3 buoy now fails closed by name, like Rigid6 bodies and rods.
  • Fold note for EI = 0 lines. A current toward the anchor or a falling seabed can carry a grounded run past its anchor, where it folds back with zero tension: a 2 m/s current carried a chain 20 m past its anchor, a 5° slope 24 m, without a word.
    • The equilibrium is kept; one note per line names the fold (distance past the anchor, or a turn above 150° at a deck node), its cause (current, slope, or slack run) and the remedy that fits the deck's static friction.
    • The finite-EI note on a run past its anchor now names cause and remedy by the same rule, and both skip a run that staticsFriction asLaid itself lays past its end.
    • Printed on every static route and in the OpenFAST initialisation report, which now also prints the finite-EI cable notes. No example deck prints it, and no result changes.
  • Deck numbering. The fold note and the run audit's spike warning number nodes and segments from End A, as the node channels and .Line<L>.t.out do (the spike warning counted from End B). No result changes.
  • Sub-step estimate. The point seabed contact is linearly implicit in the point step, so it sets no step limit; the manual states that the explicit object sub-step estimate leaves it out, as it does the implicit seabed damping and friction.

Tests

  • test_point_seabed_contact: a 50 t clump rests within its footprint's penetration with the footprint reaction balancing its weight, in the static equilibrium, the monolithic and staggered marches and the OpenFAST aggregate step, on a flat and a sloped structured seabed; a Point3 buoy of the same mass stays at rest in the marches and the aggregate step.
  • Fold-note decks: current, slope (flat and bathymetry), turn, frictional and as-laid cases, with check_fold_layout.py verifying the layouts; the finite-EI hairpin deck.
  • Full suite (Windows, gfortran, Release, slow tier included), Python, pre-commit run --all-files and the Sphinx build with -W pass.

Checklist

  • pre-commit run --all-files passes
  • ctest --test-dir build --output-on-failure passes
  • python -m pytest tests python/tests passes (if Python or validation records changed)
  • VALIDATION.md updated if a validation result changed (none changed)
  • CHANGELOG.md updated under [Unreleased] (heading added if missing) for user-visible changes
  • Documentation in doc/ updated for changed inputs, options, or outputs
  • Conventional Commit title (feat:, fix:, docs:, ...)
  • Commits are signed off (git commit -s) under the
    Developer Certificate of Origin

Connect/Free points (clump weights, buoys) had no seabed contact of
their own: a 50 t clump on polyester ended 2.8 m below the seabed,
carried only by the penalty of its line-end nodes, silently. A deck
Free/Connect point now rests on the seabed with the line contact law
(stiffness kBot A, downward-only damping cBot A, the C1 touchdown
blend, along the surface normal on a slope) over the footprint
A = V^(2/3), V its Vol or Mass/7850 when Vol is 0 (OrcaFlex's 3D-buoy
default for a cube); a massless junction has none. Statics, the
monolithic multibody residual, the staggered/OpenFAST/C API point step
and the point/body static equilibrium use the same law, so a grounded
point starts at rest. Penetration: 50 t clump 2.81 -> 0.95 m, 20 t on
chain 1.02 -> 0.67 m. Every deck without a grounded point is
byte-identical (1024 of 1031 decks); clump_weight_free_point.dat now
rests 4.7 cm deep instead of 6.6 cm and its AnchTen falls from 186 to
92 kN, the lines no longer holding the clump up.

EI = 0 lines converged silently to folded layouts (a frictionless line
pushed 20 m past its anchor by a 2 m/s current, a run sliding down a
slope). A note now names the fold, its cause and remedies, once per
line, on the static, TMax = 0, point-system and multibody routes and in
the OpenFAST aggregate's line report; no example or test deck triggers
it. Results do not change.

Signed-off-by: Jae Hoon Seo <jaehoon.seo@inha.ac.kr>
…s the static friction

A Point3 buoy steps as the Free point of its Body<N> attachment with the BODY Mass and Vol, but
that point kept its body type and so had no seabed contact. It now carries the same footprint
contact as a Free/Connect point, A = V^(2/3) with V = Vol or Mass/7850, in the staggered and
monolithic marches and the coupled point step of the OpenFAST aggregate and the C API, with
the same contact law and Jacobian. The static solve still does not place a Point3 buoy.

The EI = 0 fold note now bases its remedy on the deck's static seabed friction and on what the
route accepts: frictionMu against a current, staticsFriction asLaid on a slope (with the
default policy the static solve applies friction against a deck current only), and a shorter
line where friction already acts or the route rejects it. A run that staticsFriction asLaid
lays past its end along layAzimuth is not noted. The cause names a deck current acting
anywhere on the line, not only at the fold node, and a slope steeper than 0.5 degrees. The
past-the-end test needs a horizontal span of at least a tenth of the water depth.

The OpenFAST initialisation report prints each line's static-initialisation notes, so the fold
note and the finite-EI cable notes reach OpenFAST users. The docstrings of the line summary
note and CD_AGG_GetInitLine now describe both kinds of note.

The footprint is described as CableDyn's own area-equivalent of a compact body, with no claim
about another code's defaults. The options and driver-format pages name the point footprint
for cBot as well as kBot. The header of clump_weight_free_point.dat says that the clump now
rests on the seabed, with the chain carrying half its net weight.

Tests: a Point3 buoy set at the clump's static position stays at rest in the staggered and
monolithic marches and the aggregate step (it drifts 2.8 m without its contact), the clump
rests on a slope in the monolithic march as well, a buoy tethered clear of the seabed sees
no contact, and the fold note stays silent on a current deck with frictionMu, a slope with
as-laid friction, a run laid past the anchor by layAzimuth, and two lines through a grounded
clump.

Signed-off-by: Jae Hoon Seo <jaehoon.seo@inha.ac.kr>
The finite-EI note for a grounded run carried past its anchor named
"seabed friction or a shorter line" as the remedy. Seabed friction acts
in the static solve only with staticsFriction asLaid, so the note now
names frictionMu with staticsFriction asLaid, or only a shorter line
when asLaid is already on. With asLaid and a layAzimuth that lays the
run from the anchor away from the fairlead, the deck itself lays the
run past the anchor and no note is made, as for the EI = 0 fold note.

Signed-off-by: Jae Hoon Seo <jaehoon.seo@inha.ac.kr>
The finite-EI note on a run carried past its anchor and the EI = 0 fold note now take their
cause and remedy from one helper. The cause is a deck current on the line, a seabed steeper
than 0.5 degrees at the fold, or a slack run. The remedy follows the deck's static friction
and what the route accepts. The finite-EI static solve applies frictionMu against a deck
current under staticsFriction none, so frictionMu answers a current and staticsFriction asLaid
a slope; a shorter line remains where that friction already acts or the route rejects it
(frictionMu beside Connect/Free points or Point3 buoys, asLaid on the multibody march). The
static-only line solve, reached without dtM and TMax, carries no friction at all, so its note
says that friction needs dtM and TMax. This replaces the remedy of the previous commit, which
assumed that friction acts in the static solve only under asLaid.

Both notes now share one as-laid rule: no note is made when the as-laid reference itself, the
line laid from its lower end along layAzimuth (by default toward the other end) over its
length, reaches past the grounded end by more than the note threshold. This covers a line laid
from the anchor away from the fairlead and a line laid from a lower fairlead toward the anchor
and beyond it; the finite-EI rule ignored the laying end and the EI = 0 rule only looked at it.

Formal OpenFAST linearisation of a deck with a Point3 buoy now fails closed by name, like a
deck with Rigid6 bodies or rods, because the quasi-static re-solve holds the buoy at its deck
position.

A new test checks the layouts behind the no-note tests: the frictionless current carries the
run past its anchor, frictionMu holds it short of the threshold, and the as-laid run does lie
past the anchor. A finite-EI current hairpin checks the shared cause and remedy, and the
static-only slope test checks the new remedy. The docstring of the Point3 test now gives the
footprint reaction and the measured sinking without it.

Signed-off-by: Jae Hoon Seo <jaehoon.seo@inha.ac.kr>
…e sub-step estimate

On top of dev's run audit, deck-node channels and explicit object
sub-step estimate:

- The fold note of an EI = 0 line names the node of a turn by its deck
  number (1 = End A, the <J> of the node channels; EI = 0 lines are
  never refined), and the manual says so. A new deck folds a slack
  chain between two grounded ends in a current (180 deg at node 49):
  driver_ei0_fold_note_turn checks the note, and
  driver_ei0_fold_layouts checks that node 49 is the deck node of the
  fold in the static layout.
- The run audit's spike warning numbered its segment in internal order
  (End B first); it now counts from End A, as .Line<L>.t.out and the
  fold note do. No result changes.
- The manual states that the seabed contact of Free/Connect points and
  Point3 buoys is linearly implicit in the point step (stiffness and
  damping in the effective mass, like the end blocks of the lines), so
  it sets no step limit and the explicit object sub-step estimate
  leaves it out, as it leaves out the implicit seabed damping and
  friction.

Signed-off-by: Jae Hoon Seo <jaehoon.seo@inha.ac.kr>
@SMI-Lab-Inha

Copy link
Copy Markdown
Owner Author

@codex review

@chatgpt-codex-connector

chatgpt-codex-connector Bot commented Oct 20, 2026 •

Copy link
Copy Markdown

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review Status Commit Review trigger
📝 Code Review ✅ Completed 2026-10-20T05:30:09.097298Z d8bfeb5 Manual request
ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

@read-the-docs-community

Copy link
Copy Markdown

Documentation build overview

📚 CableDyn | 🛠️ Build #35260872 | 📁 Comparing d8bfeb5 against latest (f27c5e3)

  🔍 Preview build  

29 files changed · ± 29 modified

± Modified

@chatgpt-codex-connector

Copy link
Copy Markdown

Codex Review: Didn't find any major issues. 👍

Reviewed commit: d8bfeb524e

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

@SMI-Lab-Inha
SMI-Lab-Inha merged commit b8594fd into dev Oct 20, 2026
15 checks passed
@SMI-Lab-Inha
SMI-Lab-Inha deleted the fix/points-seabed-folds branch October 20, 2026 07:05
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant