1 · Executive Summary
- Yes, non-hurricane Caribbean seas can threaten cable slack — but not mainly the way you guessed. The classic 15–20 ft winter swell (North Atlantic swell episodes, cold-front “Northers,” tropical storms below hurricane strength) produces low differential load because your 44-ft footprint samples nearly the same part of a long wave, and the boat follows it. The genuinely dangerous regime is short, steep wind-sea chop at 3.5–5 seconds: your heave natural period computes to ≈ 3.3 s (with added mass), which lands almost exactly on the wave period that maximizes elevation difference across a 44-ft spacing (T ≈ 4 s, λ ≈ 2× your leg spacing).
- Slack is prevented by pretension, not by luck. With the pretensions recommended below, no cable goes slack below roughly Hs ≈ 18–20 ft wind-sea equivalent loading. Without deliberate pretension, even 6–8 ft near-resonant chop could transiently unload cables.
- The “spring” is mandatory, not optional. Bare steel cable is so axially stiff (≈ 50,000 lb/ft for a 28-ft ⅝″ span) that a modest snatch impact converts into a 12,000+ lb spike. An inline elastomeric compensator cuts that same event to ≈ 4,600 lb. The spring is what lets you use ⅝″–¾″ duplex instead of ⅞″–1″ rope.
- Recommended hardware: 4 cables per leg of ⅝″ duplex (2205) 7×19 wire rope, mean tension ≈ 4,500 lb each — or 3 cables per leg of ¾″ if you prefer fewer parts. Top-mounted progressive elastomeric compensators, ≈ 750 lb/in class, ≥ 10 in stroke, with instrumented clevis pins for load telemetry.
- Capability estimate: unrestricted operations to Hs ≈ 8–10 ft any heading; ≈ 12–16 ft with active heading control; ≈ 16–22 ft bow-on under para-anchor with the dinghy recovered and decks secured. Survival design envelope ≈ Hs 25 ft. Beyond that is hurricane territory — leave.
- Your instinct about diagonal seas is correct: the governing load case is “two legs lifted, one dropped,” which diagonal/quartering seas maximize. A bow sea anchor converts diagonal seas into head seas and is the single cheapest capability upgrade.
2 · Design Basis & Assumptions
Your description fixes the geometry but not the exact cable topology, so the numbers below use a stated baseline. Every result scales with the formulas given — substitute your real values as the design freezes.
| Parameter | Value used | Note |
|---|---|---|
| Full-load displacement Δ | 27,500 lb (12.5 t) | Your rated buoyancy at waterline |
| Static buoyancy per leg B₀ | ≈ 9,170 lb | Equal thirds assumed |
| Waterplane area per leg Awp | ≈ 20 ft² | Back-calculated from your “1 ft = 1/7 of buoyancy” (27,500/7 ÷ 64 psf) |
| Heave stiffness k | ≈ 3,840 lb/ft (3 legs) | ρg·ΣAwp |
| Heave natural period Tn | ≈ 3.0 s bare; ≈ 3.3–3.5 s with added mass | m = 854 slugs; entrained water adds ≈ 230 slugs for three vertical foils |
| Cable baseline | 3 or 4 cables per leg, ≈ 40° from vertical, fanning to frame hard points | Leg top bears (compression); cables restrain and carry share of load (tension) — adjust for your actual topology |
| Leg radius from centerline r | ≈ 25.4 ft | Triangle circumradius, 44/√3 |
| Cable length L | ≈ 26–30 ft | Affects stiffness & pluck-test frequency |
Key derived fact: your heave Tn ≈ 3.3 s sits inside the 3.5–5 s band where short Caribbean trade-wind chop lives and where the geometric differential across 44 ft is near maximum. Your own heave plates (good — they add damping and added mass) and the inline springs are the two features that tame exactly this regime.
3 · Can Non-Hurricane Caribbean Waves Slack a Cable?
3.1 How big do non-hurricane Caribbean seas actually get?
| Source | Where/When | Typical Hs | Extreme individual waves |
|---|---|---|---|
| Trade-wind seas | Throughout, year-round | 3–8 ft, T 5–7 s | ~10 ft |
| Winter “Northers” (cold fronts) | Bahamas, Turks & Caicos, Florida Straits, Nov–Mar | 10–15 ft, T 6–8 s | 18–20 ft |
| North Atlantic swell episodes | North-facing shores of PR, DR, Bahamas, Leewards, Dec–Mar | 10–15 ft, T 12–14 s | 20+ ft faces at exposed spots |
| Tropical storms (< 74 mph) | Jun–Nov, anywhere | 15–25 ft | 30 ft |
| Squall lines / short fetch chop | Anywhere, brief | 4–8 ft, T 3–5 s | ~9 ft |
Correction to your premise: 15–20 ft seas do occur in the Caribbean without hurricanes — winter extratropical systems and just-below-hurricane tropical storms produce them. Plan for them; don’t assume hurricane-only.
3.2 The differential-load table (the number that actually matters)
For two points spaced d = 44 ft apart in a wave of height H and length λ, the elevation difference amplitude is:
Δηgeo = H · sin(π·d/λ) λ = 5.126·T² (deep water)
| Period T (s) | λ (ft) | sin(πd/λ) | Character |
|---|---|---|---|
| 3.0 | 46 | 0.14 | Spacing ≈ 1 wavelength — legs in phase |
| 3.5 | 63 | 0.81 | Danger band: large differential and near heave resonance |
| 4.0 | 82 | 0.99 (max) | |
| 5.0 | 128 | 0.88 | |
| 6.0 | 185 | 0.68 | Trade-wind sea — significant differential |
| 7.0 | 251 | 0.52 | Moderate |
| 8.0 | 328 | 0.41 | Moderate |
| 10.0 | 513 | 0.27 | Your intuition starts to hold here |
| 12–14 | 740–1000 | 0.14–0.19 | Winter swell — legs see nearly the same water |
Your statement “normally the bigger the wave, the more all floats feel the same part of the wave” is true only for T ≳ 9–10 s. A 12-ft sea at 5.5 s has 3–4× the differential of a 20-ft swell at 13 s. Height is not the hazard metric — period structure is.
3.3 Translating differential elevation into cable load
ΔFleg ≈ ρg · Awp · ηrel (per leg, vertical)
ηrel = ηdiff,geo × (1 − following efficiency) — following efficiency ≈ 0.6–0.8 for T ≥ 6 s, ≈ 0.2–0.5 near Tn
ηrel = ηdiff,geo × (1 − following efficiency) — following efficiency ≈ 0.6–0.8 for T ≥ 6 s, ≈ 0.2–0.5 near Tn
| Sea state | ηrel (ft) | ΔF per leg (lb) |
|---|---|---|
| Hs 6 ft, T 3.5 s (near-resonant chop) | ≈ 2.0 | ±2,600 |
| Hs 8 ft, T 5.5 s (fresh trade wind sea) | ≈ 3.2 | ±4,100 |
| Hs 13 ft, T 9 s (winter front) | ≈ 2.6 | ±3,300 |
| Hs 15–18 ft, T 6.5 s (severe non-hurricane wind sea) | ≈ 4.0 | ±5,100 |
| Hs 20 ft, T 13 s (big winter swell) | ≈ 1.2 | ±1,500 |
Verdict: Adopt a design differential of ΔFdesign = ±5,000 lb per leg (vertical), with a ×2 asymmetry factor on the worst-loaded cable of each leg (pitch/roll bias, installation tolerance, the “two legs up, one down” case). This covers everything the non-hurricane Caribbean reliably produces, including the near-resonant chop that your heave Tn makes special. With the pretensions in §6, cables stay positively tensioned throughout this envelope.
4 · Slack Physics and the Snatch Calculation
4.1 The slack criterion
T(t) = Tmean + ΔT·sin(ωt) → slack if Tmean < ΔT
Design rule: Tmin = Tmean − ΔT ≥ 0.15 · Tmax,working
Design rule: Tmin = Tmean − ΔT ≥ 0.15 · Tmax,working
The 15% floor keeps fittings bedded, eliminates micro-slip fretting at terminations, and guarantees the cable never sees compression events (which destroy wire ropes quickly).
4.2 Why bare steel cable makes snatch so violent
Axial stiffness of the rope itself: k = AmE/L. For ⅝″ 7×19 (metallic area ≈ 0.175 in², E ≈ 8.5×10⁶ psi, L = 28 ft):
krope ≈ 4,400 lb/in ≈ 53,000 lb/ft (!)
Snatch peak for an impact of velocity v and effective mass m (leg steel ≈ 2,500 lb + entrained water ≈ 3,000 lb → m ≈ 170 slugs; v = 4 ft/s):
Fpeak = T₀ + √(k·m·v²) = T₀ + √(53,000 × 170 × 16) ≈ T₀ + 12,000 lb
With the same impact taken by a 750 lb/in elastomer in series (combined k ≈ 7,700 lb/ft):
Fpeak = T₀ + √(7,700 × 170 × 16) ≈ T₀ + 4,600 lb (stroke used ≈ 7 in)
This single calculation justifies the entire spring program: a ~60% reduction in snatch peak, which is the difference between sizing ⅝″–¾″ duplex versus ⅞″–1″ rope, and between 10-year and 3-year fitting fatigue life.
Honest nuance: the spring does not reduce slow, wave-forced quasi-static loads (those are set by buoyancy and geometry, not stiffness). Its jobs are (a) absorbing impact/resonant energy, (b) keeping tension positive through fast transients by returning stored energy, (c) slashing fatigue ranges at terminations, and (d) giving you a calibrated deflection-to-load readout.
5 · Inline Compliance (“Spring”) — Options and Specification
5.1 Placement: top of the cable, at the frame — confirmed
Your four reasons are all valid. Three more:
- Away from the slam zone — the lower cable region takes spray and wave slap; elastomers and calibration marks hate that.
- Gravity drainage — mount with a drip loop below the spring so water never ponds in the spring core or pin bores.
- Sensor proximity — cameras/load pins sit next to the wiring loom and computers instead of needing a wet-mate conduit down the leg.
Protect the top unit from UV with a fabric or HDPE cover, and from dropped-object strike with a simple bail.
5.2 Option comparison
| Option | Snatch absorption | Damping | Creep / drift | Inspectability | Life | Cost/Weight | Verdict |
|---|---|---|---|---|---|---|---|
| 1) Elastomeric mooring compensator (bonded rubber, progressive) | Excellent (progressive rate + hysteresis) | Inherent, ζ ≈ 0.15–0.3 | Low–moderate (set over years) | Visual + durometer; calibrate scale on housing | 4–6 yr replace | Moderate / 15–30 lb | Primary recommendation |
| 2) Nylon rope section (8-plait, 4–6 ft) | Very good (huge stretch) | Good (hysteresis) | High — creep + wet/dry stiffness swings; pretension drifts ±20% | Poor — load invisible; chafe hidden in strands | 2–3 yr | Cheap / light | Budget fallback or sacrificial element in series; demands frequent retension |
| 3) Metal coil spring (duplex/17-4PH wire) | Good (linear) | None — must add a damper; spring surge under impact | Virtually none | Excellent — measure free length | 20+ yr | Moderate / 30–50 lb | Best where zero-drift matters; add elastomer washers in parallel for damping; needs catch-sleeve fail-safe against fracture |
| 4) Oleo-pneumatic strut (gas-over-oil accumulator) | Excellent, tunable | Adjustable orifice | Gas temperature effect only | Pressure transducer = direct load telemetry | Seals 2–3 yr | High / 40–80 lb | Premium option; uniquely satisfies your “sense the load” goal and permits computer-trimmed stiffness underway vs. parked |
5.3 Recommended specification (per cable, top-mounted)
| Parameter | Specification |
|---|---|
| Type | Progressive-rate elastomeric compensator, bonded natural-rubber compound, ozone/UV resistant, 65–75 Shore A, steel end caps with retention rod (fail-safe: on bond failure the rod catches at hard-stop, no dropped load) |
| Rate | ≈ 500 lb/in to 4,000 lb, rising to ≈ 1,200 lb/in to 12,000 lb (progressive) |
| Stroke | ≥ 10 in at 12,000 lb; hard-stop at 13–14 in |
| Ultimate | ≥ 20,000 lb (well below cable MBL so the cheap part is the fuse) |
| Calibration | Engraved scale on housing: deflection ↔ load, ±5%; photograph monthly |
| Environment | −10…+60 °C; drip loop below; UV cover; replace at 4–6 yr or any bond crack/bulge/set > 10% |
| Upgrade path | Oleo-pneumatic unit with 0–5,000 psi transducer → live load per cable on the boat network; enables active tension trimming with your thruster-control computers |
Hybrid worth considering: elastomer on top + a 2-ft 8-plait nylon “sacrificial tail” just above the lower shackle. The tail is cheap, chafe-visible, and replaced after any major event — it protects the expensive compensator from shock overload.
6 · Cable Specification — Duplex Stainless
6.1 Load budget (per cable)
| Quantity | 3 cables/leg | 4 cables/leg |
|---|---|---|
| Static share (axial, 40° from vertical) | ≈ 4,000 lb | ≈ 3,000 lb |
| Added pretension (turnbuckle) | +2,000 lb | +1,500 lb |
| Mean tension Tmean | ≈ 6,000 lb | ≈ 4,500 lb |
| Dynamic swing (ΔFdesign ±5,000 lb/leg, ×2 worst-cable asymmetry) | ∓4,350 lb | ∓3,265 lb |
| Tmin (slack margin) | ≈ 1,650 lb ✓ | ≈ 1,235 lb ✓ |
| Tmax quasi-static | ≈ 10,350 lb | ≈ 7,765 lb |
| Tmax with snatch (DAF 1.4, enabled by springs) | ≈ 14,500 lb | ≈ 10,900 lb |
| Required MBL at SF = 4 | ≈ 58,000 lb | ≈ 43,500 lb |
| Selected rope | ¾″ duplex 7×19 (MBL ≈ 58 klb, SF ≈ 4.0) | ⅝″ duplex 7×19 (MBL ≈ 41 klb, SF ≈ 3.8–4.0) |
Without inline springs, snatch DAF reaches 2.5–3 and both columns jump a rope size (⅞″–1″). The springs pay for themselves in weight, cost, and fatigue life.
6.2 Approximate breaking loads — verify against mill certificates
| Rope dia (7×19) | 316 SS MBL (lb) | Duplex 2205 MBL (lb, est.) |
|---|---|---|
| ⅜″ | ~12,500 | ~16,000 |
| ½″ | ~21,000 | ~27,000 |
| ⅝″ | ~32,000 | ~41,000 |
| ¾″ | ~45,000 | ~58,000 |
Duplex (UNS S32205) runs roughly 25–35% stronger than 316 and has far better chloride-pitting resistance (PREN ≈ 34). Figures are planning estimates — procure against certified test reports.
6.3 Construction, terminations, and corrosion rules
- Construction: 7×19, right regular lay, for flexibility over thimbles. Avoid 1×19 (too stiff for this duty).
- Terminations: swaged studs or resin/poured open sockets. Never wire-rope clips on a primary tensegrity member (turnback losses of 10–20% plus slip-under-cyclic-load). Clevis/rod-end fittings at both ends so misalignment never bends the rope under load.
- Bend radius: any fairlead, thimble, or deviation D/d ≥ 12; duplex is less ductile than 316 and punishes tight bends.
- Galvanic isolation: if the frame is aluminum, duplex-to-aluminum contact in a salt environment eats the aluminum. Isolate every contact with UHMW/Delrin bushings and washers, and confirm your leg anode scheme covers the splash zone.
- Crevice control: duplex resists pitting but not stagnant crevices — open sockets with drain paths, no tape-wrapped joints, hose-clamp-free attachments below deck level.
7 · Wave-Handling Optimization and Capability Estimate
7.1 Capability ladder
| Condition | Hs | Mode | Governing check |
|---|---|---|---|
| Routine, any heading | ≤ 8–10 ft | Free running / station-keeping | Comfort, dinghy drag |
| Caution, any heading | 8–12 ft | Heading freedom reduced | Cable swing margin |
| Head/quartering with active heading (thrusters) | 12–16 ft | Powered bow-on | Thruster authority vs. windage |
| Bow-on under para-anchor, dinghy recovered, decks secured | 16–22 ft | Drifting, bow to seas | Breaking-crest slam on foil noses; green water over walkway begins ≈ Hs 17–18 ft (deck ≈ 10.75 ft above WL, crest ≈ 0.67·Hs) |
| Survival design envelope | ≈ 25 ft (individual 35+) | Drifting bow-on, everything stowed | Margins thin; cable Tmin approaches zero in the worst phasing — acceptable as a survival bound, not an operating one |
| Hurricane | 30 ft+ | Not designed | Evacuate or run before the season curve |
7.2 Diagonal seas — your identified danger case, quantified
With three support points, the critical case is any event that lifts two legs while dropping the third; quartering/diagonal seas maximize both its probability and its magnitude (combined pitch+roll excitation, torsion into the frame). The ×2 worst-cable asymmetry factor in §6 exists precisely for this case. Mitigations, in order of value:
- Bow sea anchor — a 25–40 ft para-anchor on 300+ ft of nylon rode with a 40–60 lb chain rider, bridled to two forward points. Converts diagonal seas to head seas passively, no power needed. Deploy early (Hs ≈ 10 ft and building), not late.
- Active heading with your rim drives — you already plan coordinated multi-thruster control for walkway ops; extend the same controller to hold heading in seas up to thruster-authority limits, then hand off to the para-anchor.
- Active load balancing — with load pins per cable (§5.3), the computers can bias differential thrust to pre-load the “going light” cables during long wave groups. Cheap software, real fatigue-life payoff.
- Optional fourth cable per leg — redundancy plus lower per-cable loads in exactly the asymmetric case.
- Storm hardening checklist — recover the dinghy (a towed RIB in 15-ft seas is your most vulnerable component), dog the doors, latch walkway grating panels, stow anything on the roof.
7.3 Parked-mode (helical screw tension legs) note
Your 3-ft pull-down is sound for the intended protected, micro-tidal sites: slack requires trough + tide + set-down exceeding 3 ft. Two additions: (a) give the mooring motors an automatic “add pull-down” trigger when the load pins or a wave sensor report long-period swell; (b) remember that in parked mode the structural cables see the same differential game as underway — the pretension budget above already covers it.
8 · Tension Management Over Time
8.1 Why tension drifts
- Bedding-in: new rope and terminations permanently stretch ~0.5–1% in the first weeks of load. Expect a one-time pretension loss; re-tighten at 1 week and 1 month after commissioning.
- Creep: significant for any nylon in the load path (§5.2); negligible for steel and bonded elastomers below 30% rating.
- Temperature: steel + elastomer column changes length with seasons; a ±20 °F swing moves tension a few percent. Log it, don’t chase it.
- Wear and set: elastomer set, socket wedge settling, shackle bore ovalization.
- Events: any snatch, slam, or storm above the design sea state warrants a full re-survey.
8.2 Measuring tension (three ways, cheapest first)
- Compensator deflection: read the engraved scale — this is why the spring lives at the top where you can see it.
- Pluck test (vibration method): strike the cable, read the fundamental frequency f₁ with a phone accelerometer or clip-on piezo:
T = 4·μ·(L·f₁)² Example: ⅝″ duplex, μ ≈ 0.75 lb/ft, L = 26 ft, measured f₁ = 8.4 Hz → T ≈ 4,500 lb ✓
- Instrumented clevis pins (strain-gauged, telemetered) — the gold standard, doubles as your continuous load-monitoring system and feeds the active-balancing controller.
8.3 Adjustment procedure
- Target: every cable within ±5% of design mean; Tmin never below 15% of Tmax,working.
- Adjust in a star pattern around each leg (opposite pairs first), small increments, re-measuring after each pass — never full-tension one cable while neighbors are slack.
- Re-check at 1 week, 1 month, then quarterly and after any extreme event. Keep a dated log per cable (tension, deflection, temperature, notes).
- If a cable repeatedly loses tension faster than its siblings, suspect termination bedding or a fatigued strand — inspect, don’t just re-tighten.
9 · Fatigue, Inspection, Cleaning
9.1 Fatigue drivers and budget
Wire-rope fatigue scales with mean tension and tension range. Your duty cycle: trade-wind chop at 5–6 s delivers ~14,000 cycles/day (~5M/yr). With the §6 sizing, the everyday range is ±6–8% of MBL — comfortably inside the multi-million-cycle regime for 7×19 rope with quality terminations. The fatigue hot-spots are terminations and bend points, not the rope body: hence the D/d ≥ 12 rule, rod-end fittings, and the 15% minimum-tension floor (which eliminates the fretting micro-slip that occurs near zero tension).
9.2 Inspection schedule
| Interval | Actions |
|---|---|
| Monthly | Visual along full rope: broken wires, kinks, necking, corrosion bloom; compensator: cracks, bulges, set; photograph scale readings; verify turnbuckle lock nuts |
| Quarterly | Fresh-water wash of all cable and fittings; measure spring free-length/deflection vs. log; check fairlead liners and UHMW isolation bushings |
| Annual | Full tension survey (pluck or load pins); caliper the rope at 3 stations (replace at ≥ 10% diameter loss); dye-penetrant or magnflux on lugs/pins; elastomer durometer check; diver or haul-out check of lower third and splash zone |
| Every 2–3 yr | Replace nylon tails/sacrificial elements; seal service on any oleo struts |
| Every 4–6 yr | Replace elastomeric compensators regardless of appearance |
| Every 8–10 yr | Replace all primary cables on calendar life, condition notwithstanding |
| Post-event | Any snatch/slam/storm beyond design: full inspection before restoring full pretension |
9.3 Discard criteria (any one triggers replacement)
- ≥ 6 broken wires in one lay length (7×19), or any valley break
- Kink, crush, bird-cage, or heat discoloration
- Corrosion pit deeper than 10% of wire diameter; rust staining bleeding from inside the strand (internal — the dangerous kind)
- ≥ 10% diameter loss at any station
- Compensator: bond crack, bulge, permanent set > 10%, or hardness shift beyond spec
9.4 Cleaning and fouling
Fresh-water rinse quarterly (salt retention drives crevice corrosion at fittings). The submerged lower thirds will foul — soft growth is cosmetic; reject hard-shell accumulation at annual inspection since it hides the rope surface. Light internal lubrication at installation only (open sockets drain; do not grease-pack compensators). During haul-outs, let the lower cable sections dry fully before inspection — wet wire hides broken wires.
10 · Replacement With Dual Attachment Points
Your plan — two rated lugs at each end so the new cable is rigged before the old is cut — is exactly right. The subtlety is load transfer: two parallel cables of different stiffness do not share load predictably; a newly tensioned twin may carry almost nothing until the old one is nearly unloaded, then pick it up abruptly.
10.1 Procedure
- Rig new cable slack on the second lug pair, with its own turnbuckle fully backed off. Safety-pin everything.
- Measure the old cable’s tension (scale/pluck/load pin). Record it.
- Alternate in small steps: back off the old turnbuckle ~10% of its tension, then bring the new turnbuckle up by the same amount. Re-measure both. Repeat. The goal is to hold total path tension constant so the leg never feels the swap.
- Below ~10% of design tension, the old cable is safe to detach: secure it with a temporary grip/sieze before opening any fitting — a 5,000 lb pretension stores roughly ½·T·δ ≈ 1,500–2,000 ft-lb in a 28-ft span. That is a lethal amount of stored energy; nothing gets cut under load, ever.
- Remove old cable, inspect its lugs/pins (this is your chance to NDT the hardware), then either leave the freed lugs as the next generation’s spare path or move the new cable to the primary position in a second staged transfer.
- Bedding pass: after 1 week, re-check the new cable’s tension (it will have lost a few percent to bedding) and top it to design mean.
Stored-energy rule: treat every tensioned cable as a loaded spring. Control release only through the turnbuckle, keep faces out of the plane of any fitting, and never stand in line with a socket.
Design detail worth adding: make both lugs at each end identical and fully rated (not one “primary” and one “spare”), and orient the spare hole so the parallel cable clears the compensator housing and walkway structure.
11 · Caveats and Next Steps
- All load figures derive from the stated baseline (§2). When your cable topology, angles, and leg mass are frozen, rerun §4–§6 with the formulas provided — the method transfers, the numbers will shift.
- The heave Tn ≈ 3.3 s finding deserves a model test or at least a validated seakeeping run (e.g., a strip-theory or CFD diffraction analysis of the three-foil geometry) before finalizing pretension. It is the pivotal dynamic parameter in this design.
- Have the frame lug weldments and pin bores checked by a structural engineer against the 14,500 lb ultimate cable load with appropriate weld factors.
- Procure rope against certified breaking-test certificates, not catalog numbers — duplex batch variation is real.
- Consider a one-day instrumented trial: load pins + accelerometers in 6–10 ft seas will validate the entire load model within your first month of operation, and the data stream feeds the active-balancing controller you already planned.