Investigation 1 · TEWS-A3opened 12 Oct 2026, 10:02 by K. Venkat3 plan versions · 5 Runs · 1 branch · 1 fork
Anchor and mooring of tsunami buoy A3: a suction caisson in soft clay at 3,850 m that survives a Mw 9 earthquake, a turbidity current and a 100-year cyclone
Open · 1 Run running
KV
MK
AR
7.1 of 10 CU reserved · 5.9 spent · 2.9 left
Needs you
G3 Anchor design waiting on your sign-off · due 18:00 · branch Ø4.0 × 8.0 m · Run 5 · then A. Rao
1 result stale since polyester certificate v2 · line utilisation and fatigue, Run 4 step 14, carried into Run 5
Elidyne drafts a plan version that applies certificate v2 and shows its diff here. Nothing runs until you approve it.
Monday 12 October 2026
K. Venkat10:02
Plan the anchor and mooring of buoy A3 from the attached core, catalogue, hindcast and drawing. It has to survive the Mw 9 it exists to detect, then a cyclone on the softened clay, and keep detecting. Start with the site and the hazards; add capacity and installation once the site basis is through. Flag anything you would have to assume.Workbench
GC-7_core_log.pdfvane su · 9.4 mregional_seismic_catalogue.csv1,284 eventscyclone_hindcast_40yr.nc40 yearsTEWS-A3-S-02_caisson.pdfrev Apolyester_certificate.pdfv1
E
Elidyne10:03Workbench
I read the core log, the catalogue, the hindcast, the caisson drawing and the line certificate. Station A3 sits in 3,850 m of water in the Andaman Sea, 212 km west of the Sunda trench. Core GC-7 gives soft pelagic clay, su = 3 + 1.4z kPa with sensitivity 3. Drawing S-02 rev A shows a steel suction caisson Ø3.5 × 7.0 m, 14 t, holding an inverse catenary of 120 m chain, 4,300 m polyester, 3.2 t of buoyancy and 60 m of chain under a 2.6 m discus buoy of 4.2 t.
The design needs inputs these files do not give: . Two of them stop the analysis, the design earthquake and a turbidity current velocity. The catalogue lists three megathrust segments within 400 km but names no design event, and nothing in the files measures a turbidity current, so I have not picked either. Until both are settled I can plan the site model only.
Design earthquake: segment, magnitude and distance
Turbidity current velocity at the site
Cyclic strength tests on GC-7
Clay unit weight below 6 m
Padeye detail and shackle rating
Buoy drag and added-mass coefficients
Polyester dynamic stiffness
Seabed slope at the anchor position
Installation vessel pump and ROV limits
K. Venkat10:14
Design for a Mw 9.0 megathrust at 180 km. Take the turbidity current at 2.5 m/s as an assumption and keep it open.Workbench
E
Elidyne10:15Workbench
The design event is a Mw 9.0 megathrust at 180 km, cited to the catalogue segment you named at 10:14. Turbidity enters at 2.5 m/s as an assumption; it opens finding T-6, and every result that uses it carries the label. With both settled I can plan the site model and the hazard loads.
Plan v1 · checked · 0 holesApproved by K. Venkat · 10:20 · reserved 4.2 CU
Lanes 1–2: site and station model, hazard loads
Design for a Mw 9.0 megathrust at 180 km. Take the turbidity current at 2.5 m/s as an assumption and keep it open.K. Venkat · 10:14
9 stepsNew 92 decisions recorded7 missing inputs open · none blockingreserve 4.2 CU
1 Site and station model 4 stepsG1 Site basis2 Hazard loads 5 stepsG2 Hazard basis
Runs 1–2 partly superseded by v2
Runs 1–2 · plan v1 · from background execution
Run 1 started on meridian-hpc · OpenSees 3.7 and MoorDyn 2.3 · 8 cores · Run 2 queued for 128 cores
Lane 1 done: station model assembled · 146 elements · static equilibrium found
Gate G1 Site basis opened for M. Krishnan. Run 2 continues; its results are held until G1 is decided.
1 Site and station model2 Hazard loads · running since 11:05
Wednesday 14 October 2026
Gate G1 · Site basisChanges requested by M. Krishnan · 11:40
Reviewer M. Krishnan, geotechnics · opened 12 Oct 11:05 · records from Run 1, lane 1
as modelled: mean su and padeyeinstallation reads the tip hereSchematic, not to scale
Soil
core GC-7, 9.4 m · vane su = 3 + 1.4z kPa · sensitivity 3 · unit weight extrapolated below 6 m
Station model
146 elements · buoy 1, line 132 lumped masses, floats 4, caisson macro-element 9 · static equilibrium found
Load path
buoy → chain 120 m → polyester 4,300 m → buoyancy 3.2 t → chain 60 m → padeye → caisson → clay
Installation
penetration screen read the mean su line; plug-heave allowable taken from the mean
Missing inputs
7 open · 2 bear on this gate: cyclic strength tests, unit weight below 6 m
Changes requested by M. Krishnan · 11:40
Use the lower-bound su for installation, not the mean.
Elidyne drafted plan v2 from this note: installation reads the lower-bound line su = 2.2 + 1.2z kPa, and capacity keeps the mean line, which the note does not change. Lane 2 continued while the gate was open; its results were held until the decision.
E
Elidyne11:42from the G1 note
From M. Krishnan’s note I changed one input: installation now reads the lower-bound su line, 2.2 + 1.2z kPa, instead of the mean. Capacity keeps the mean. Three steps read the installation line and run again. Five finished steps do not, and are carried from Runs 1–2 at no charge. As you asked on Monday, I also added lanes 3 and 4, anchor capacity and installation, on the drawing’s Ø3.5 × 7.0 m caisson. The seismic site response is still running under v1; choose what happens to it.
Plan v2 · from v1 at gate G1 · checked · 0 holesApproved by K. Venkat · 12:05 · reserved 1.4 CU
Installation on lower-bound su; adds anchor capacity and installation
Use the lower-bound su for installation, not the mean.M. Krishnan at gate G1 · 11:40
Keep 5Re-run 3New 5Dropped 0In flight 1+1.4 CU
Re-run 3est. 0.14 CU
Soil design lines · Lane 1 · step 1
installation line: mean su → lower-bound su 2.2 + 1.2z kPa · from the G1 note
est. 0.02 CU
Penetration screen · Lane 1 · step 4
input changed: installation su line · self-weight and suction penetration to 7.0 m
est. 0.08 CU
Load case matrix · Lane 2 · step 9
input changed: LC5 installation reads the lower-bound line · not yet run in Run 2
est. 0.04 CU
New 5est. 1.26 CU
Capacity envelope, intact · Lane 3 · step 10
added: H–V capacity of the caisson at three load inclinations, DNV-RP-E303
est. 0.42 CU
Post-quake strength degradation · Lane 3 · step 11
added: capacity on clay softened by the Mw 9 cycles
est. 0.36 CU
Capacity checks by load case · Lane 3 · step 12
added: factor of safety for LC1–LC4 against the owner criterion FoS ≥ 2.0
est. 0.18 CU
Installation suction and plug heave · Lane 4 · step 13
added: required against allowable suction on the lower-bound line
est. 0.18 CU
Line utilisation and fatigue · Lane 4 · step 14
added: chain and polyester utilisation, top-chain fatigue life
est. 0.12 CU
Keep 50 CU · linked, not recomputed
Dropped 0
Nothing dropped.
In flight 1as settled
Seismic site response · Lane 2 · step 7
Running under v1 in Run 2 since 13 Oct 21:50 on 128 cores. It reads the mean su line, which the note does not change.
When it finishes, its fingerprint is compared with v2. A match is carried at 0 CU; anything else runs again in Run 3. Chosen at approval · finished 13:18 · matched · carried.
as settled
Cost of approval+1.4 CUre-run and new only · the 5 kept steps charge 0 CU
Runs 3 and 4 started under v2 · 5 records carried from Runs 1–2 at 0 CU · 8 steps dispatched
Seismic site response finished in Run 2 · fingerprint matches v2 · carried into Run 3 · Runs 1–2 closed: 3.5 CU spent, 0.7 released
Run 4 finished · installation and line checks · 0.4 CU · results held until G1 and G2
G1 Site basis accepted by M. Krishnan · pinned 12 records of the v2 site model
Run 3 lane 3 started · caisson capacity at 3 load inclinations on meridian-hpc
K. Venkat16:45
What governs the load on the anchor?Instant · about main · v2 · Runs 3–4
E
Elidyne16:45Instant · from records
The cyclone, pulling on clay the earthquake has already softened. The 100-year cyclone (Hs 9.4 m, Tp 13.8 s, wind 38 m/s) drives the line tension to at 32° to the seabed. The Mw 9 adds little load at the padeye but costs the clay its strength: 15 equivalent cycles at 0.28 g give . A 2.5 m/s turbidity current adds of drag on the caisson, an assumption until T-6 is closed. So the case to watch is LC3, the quake followed by the cyclone.
Based on lanes 1–2. Lane 3 (capacity) is running; about 35 min left.
While you were away · 7 days
14 Oct 16:46 to 21 Oct 11:20 · 4 updates
Run 3 finished · capacity at 3 inclinations · 0.8 of 1.0 CU · checks held until G2
Deferred answer posted · to your question at 16:45
1 result stale · D. Gupta uploaded polyester certificate v2
G2 Hazard basis accepted by A. Rao · turbidity velocity recorded as an assumption
E
Elidyne14 Oct 17:24deferred answer to your question at 16:45
Lane 3 finished. The softened clay governs, not the cyclone alone. After the Mw 9 the clay keeps 82 % of its strength, and the capacity of the Ø3.5 × 7.0 m caisson with the line at 32° falls from 392 kN to . Under the cyclone peak on that softened clay the padeye moves .
Based on lanes 1–3 of Runs 1–3 · plan v2. G2 not yet decided; the load-case checks are held.
Run 3 finished · 3 steps run · 0.8 of 1.0 CU spent, 0.2 released · load-case checks held until G2
Tuesday 20 October 2026
Polyester certificate v2 uploaded by D. Gupta · dynamic stiffness revised, MBL 1,450 kN unchanged · 1 result stale: line utilisation and fatigue, Run 4 step 14 · a revision is proposed, nothing re-ran
Wednesday 21 October 2026
G2 Hazard basis accepted by A. Rao · pinned 14 records · turbidity 2.5 m/s recorded as an assumption, finding T-6 open · load-case checks released
Completed · 1 design check failsRuns 3–4plan v2 · checks released 11:051.2 of 1.4 CU · 0.2 released
Lanes 1–4 · anchor Ø3.5 × 7.0 m from drawing TEWS-A3-S-02
14 steps · 6 carried at 0 CU · 8 run · 1 result stale
Summary
Result document excerpt · thread density
On clay the Mw 9 has softened, the drawing’s caisson does not hold with the margin you set. The quake-then-cyclone case LC3 gives a factor of safety of , below 2.0 for every one of the 20 cyclone seeds. The cyclone on intact clay passes at , and installation on lower-bound clay has a margin of .
needed capacity, peak demandSolid: intact · dashed: after the Mw 9 · LC3 demand from Run 2, carried
Key results
4 of 12 quantities · 1 fails · 1 stale
Quantity
Value
Design check
Verdict
Source
Anchor FoS · LC3 quake then cyclone1.66–1.83 over the 20 seeds
Fails · FoS < 2.0
Accepted
[6]
Anchor FoS · LC1 cyclone, intact clay2.02–2.23 over the 20 seeds
Passes
Accepted
[6]
Installation margin · plug heaveLC5 · lower-bound su · 30 of 87 kPa
Passes
Accepted
[7]
Polyester utilisation · LC1 peakcarried from Run 4 · step 14Stale · input changed: polyester certificate v2 Elidyne drafts a version that applies certificate v2: line utilisation and fatigue re-run, everything else is kept.
11 Post-quake strength degradation · cyclic contour, 15 cycles · 0.29 CUCaveated · no cyclic tests on GC-7
12 Capacity checks by load case · LC1–LC4 against FoS ≥ 2.0 · 0.18 CUAccepted
13 Installation suction and plug heave · lower-bound su · 0.17 CUAccepted
14 Line utilisation and fatigue · DNV-OS-E301 and API RP 2SK · 0.10 CUAccepted
G3 Anchor design not opened on main · LC3 is below FoS 2.0; A. Rao is asked once a size passes
Monday 2 November 2026
K. Venkat14:02
Try Ø4.0 × 8.0 m, and add the calamity sequences on it.Workbench · main · v2 · Runs 3–4
E
Elidyne14:03Workbench
I’ll branch from v2 so the main line stays as it is. A scaling screen on Run 3 puts Ø4.0 × 8.0 m near 2.7 after the quake and Ø4.5 × 9.0 m near 3.6, which carries 6 t more steel than the criterion needs. Every step that depends on the caisson size runs again; the other six are carried at no charge, and lane 5 adds the sequences in the time domain.
Branch “Ø4.0 × 8.0 m” from v2active · compare pending
in branch Ø4.0 × 8.0 m
Plan v3 · from v2 · checked · 0 holesApproved by K. Venkat · 14:06 · reserved 2.4 CU
Try Ø4.0 × 8.0 m, and add the calamity sequences on it.K. Venkat · 14:02
Keep 6Re-run 8New 2Dropped 0+2.4 CU
Polyester certificate v2 is not applied on this branch either. The kept line check still reads certificate v1, so its utilisation stays stale until you revise.
in branch Ø4.0 × 8.0 m
Run 5 · plan v3 · from background execution
Queued on meridian-hpc · 128 cores · position 1
Running · 6 records carried at 0 CU · 10 steps dispatched
Lanes 3–4 done: capacity, checks and installation · gate G3 opened on this branch
Capacity check LC3 at , held until G3
Lane 5: calamity sequences running · 14 of 20 seeds
Spent1.2 CU
Reserved2.4 CU
Investigation left2.9 CU
1 Site2 Hazards3 Capacity4 InstallationG3 waiting on you5 Sequences 14 of 20 seeds
in branch Ø4.0 × 8.0 m
Gate G3 · Anchor designwaiting on you · due 18:00
Reviewer: you, as branch owner, then A. Rao, offshore structures · opened 16:40 · records from Run 5, lanes 3–4
Solver converged at every load step · OpenSees 3.7 macro-element · 0 warnings
Equilibrium at the padeye, LC3: applied 185.0 kN at 32°, reaction 185.0 kN
Independent limit-equilibrium check, DNV-RP-E303: intact capacity within 4.1 % of the macro-element
Cyclone seeds 20 of 20: LC3 FoS 2.62–2.89; post-quake strength from correlation, no cyclic tests on GC-7
Installation on lower-bound su: suction required 38 kPa, allowable 92 kPa, margin 2.4
Lane 5 continues while this gate is open; its results are held until you decide.
K. Venkat opened Compare · main v2 with branch v3
in branch Ø4.0 × 8.0 m
Compare · main v2 with branch v3
Compare pending · 1 quantity still running
Runs 3–4 (Ø3.5 × 7.0 m) against Run 5 (Ø4.0 × 8.0 m) · aligned by quantity · G3 is open on the branch