Short answers to the three questions
- Have I seen this before? Yes — the function you describe (rigid below a force threshold, compliant above it, self-resetting) is a well-known mechanism class. It appears as ball-detent overload clutches in machinery, torsion-bar load limiters in seat belts, downrigger release clips in fishing, energy absorbers in climbing gear, and gas-over-oil riser tensioners offshore.
- Is there a name? The ball-in-socket element is a detent (a ball detent in a conical seat). The whole assembly would be called a load limiter, overload release, or — for this application — a preloaded snatch load limiter (a "snubber with a deadband"). There is no standard marine-chandlery name because off-the-shelf marine gear with a true deadband is rare; ordinary mooring springs and rubber snubbers are compliant from the first pound.
- Is it a good basic design? Yes — the preloaded spring is exactly the right way to get a deadband, and because the seated ball carries the load metal-to-metal, the spring is not in the load path during normal waves, so your lines stay effectively rigid. The details below (threshold setting, travel, rebound, end-of-travel, fouling, fail-safe) decide whether it works in the real ocean. Two cheap additions — a slack standby strop and a slip-drag mode on your existing winch motors — make it robust.
Why the snatch load gets big
With a 1-ft pull-down you are carrying roughly (27,500 lb ÷ 7) ≈ 3,900 lb of total pre-tension, about 1,300 lb per line. A wake crest larger than the pull-down margin unloads a line completely. Note that dynamically this happens early: a low-stretch line under ~1,300 lb of tension has only a percent or so of elastic strain, so the first few inches of lift take the tension to zero — after that the platform is a free-floating body with gentle net accelerations (its buoyancy-vs-weight imbalance near the free-float draft is only a few thousand pounds, i.e. a small fraction of g).
The damage is done on the way back down. The platform returns toward the tensioned position with some vertical velocity — a steep 2-ft wake can plausibly deliver 2–4 ft/s at the moment the line re-tautens (refine with a time-domain heave/pitch model; your heave plates help via damping but hurt via added mass). When a moving mass m with velocity v is caught by a spring of stiffness k:
Fpeak = v · √(k·m)
A short HMPE line plus fittings is easily 100 kip/ft or stiffer, so even a 3 ft/s catch produces a peak on the order of 20–26 kip on a line whose static load is 1.3 kip — 15–20× static. That is the load that bends padeyes, strips winch gears, or plucks a helical screw out of the seabed. The cure is always the same: add controlled travel so the energy is absorbed over distance instead of instantly. Your device adds travel only when it is needed — which is the elegant part.
Also keep in mind the second overload mode: a big trough with a fully taut line spikes tension with no slack involved. Your preloaded limiter caps that mode too; a purely "slack-activated" backup does not.
Prior art & what to call it
The mechanism principle — a spring preloaded against a seat, so nothing moves until the applied force exceeds the preload — is used across many fields:
| Field | Device | How it matches your concept |
|---|---|---|
| Power transmission | Ball-detent overload clutch / torque limiter | Balls held in detents by spring preload; they pop out at a set torque and self-reset. Your design is the linear version. |
| Automotive | Seat-belt load limiter | Retractor locks rigidly; above a set force a torsion bar yields and pays out webbing at controlled force. Rigid-then-compliant, exactly your force curve. |
| Fishing | Downrigger release clip; reel drag | Adjustable-tension release (detent-like); a reel drag pays out line above a set tension while dissipating energy — see the winch slip-drag option below. |
| Climbing | "Screamer" / via-ferrata energy absorber | Stitching rips above ~2 kN to absorb a fall; deadband plus travel, but one-shot rather than self-resetting. |
| Offshore oil | Riser tensioners (gas-over-oil) | Preloaded accumulators give near-constant tension with compliance — the industrial-scale cousin of your spring. |
| Yachting | Nylon anchor snubber, rubber mooring compensator, stainless mooring spring | Same job (killing snatch loads) but no deadband — they are soft in normal conditions too, which is what you are trying to avoid. |
Naming: call the ball-and-socket a ball detent or detent seat, and the whole box a preloaded load limiter, overload release, or snatch load limiter. If you document it as "a mooring snubber with a deadband," any marine engineer will immediately understand it.
The ball-detent limiter, done right
Design refinements that matter
- Set the threshold from troughs, not crests. The device trips on high tension. Your normal worst case is static tension plus the trough increment. Set preload ≈ 1.3–1.5× that sum, or the device will stroke (and wear) in ordinary waves.
- Guide the ball on a stem. A loose ball climbing out of a conical socket has snap-through (negative stiffness) and can side-jam. Put the ball on a stem guided in a low-friction bushing (acetal or bronze) so the unseating force equals the spring preload, crisply and repeatably. The cone seat then only centers the ball.
- Size travel from energy, not force. Required travel × average force ≥ worst-case kinetic energy at re-catch, including added water mass. See the worked example below.
- Plan for rebound. A pure spring returns nearly all the energy it stores. Your heave plates give strong system damping, so some bounce is acceptable — but an elastomer stack (rubber or polyurethane in compression) instead of a metal spring gives hysteresis damping for free, at the cost of slow creep (re-check preload seasonally). A small oil dashpot in parallel is the deluxe option.
- Never let it bottom out hard. If the spring reaches solid height, the remaining energy goes into the low-stretch line and you get the snatch anyway — with interest. Provide generous travel margin and an elastomer bumper for the last 10–15% of stroke.
- Keep it out of the sea. A bare ball-and-seat in splash will foul with growth and can literally get glued to the seat; stainless springs suffer crevice corrosion and stress-corrosion cracking in seawater. Mount the device at the corner above the waterline (your plan), in a housing with a drain and grease, with a boot over the stem. Co-locate it with the winch/motor unit so the whole tensioning station is one inspectable assembly.
- Fretting while seated. In normal waves the seated ball sees the full cyclic load metal-to-metal — millions of cycles per season. Use a replaceable polymer seat liner and generous contact radii, and grease it.
- Fail-safe backup strop. Run a second, slightly slack line (strong enough alone, with a small snubber) in parallel. If a spring breaks, a guide jams, or travel runs out, the corner is caught softly instead of going free. This also covers the fuse option below.
- Add a tell-tale. A sliding marker ring or shear-tie shows the device stroked and how far — so the owner learns the event happened and inspects. Better: a load cell logged by your onboard computer.
- Mind the line angle. Keep the line axis aligned with the spring axis (swivel or fairlead ahead of the device); side loads cause stiction and raise the effective trip force.
Worked sizing example (illustrative — not a substitute for analysis)
Numbers, using your design values
- Static: 1-ft pull-down ⇒ ≈1,300 lb per line (27,500 lb ÷ 7 per ft, ÷ 3 lines).
- Normal waves: 1-ft harbor waves swing tension roughly ±650 lb ⇒ normal max ≈ 1,950 lb. Choose preload P ≈ 2,600 lb. The ball never moves in normal waves.
- Design event: a 2-ft wake crest (1 ft over the margin). The corner free-heaves ~1 ft and the line re-catches it coming down at ≈3 ft/s (order of magnitude for a steep wake — net accelerations near the free-float draft are only ~0.1 g, so catch speeds stay in the low single digits of ft/s).
- Energy: effective mass at one corner ≈ 60% of (platform + added mass) ≈ 750 slugs ⇒ KE = ½·m·v² ≈ 3,400 ft·lb.
- Absorber: with P = 2,600 lb and spring rate k = 3,000 lb/ft, energy absorbed over travel x is P·x + ½·k·x². Setting equal to 3,400 ft·lb gives x ≈ 0.9 ft (≈11 in) and peak ≈ 5.3 kip.
- Comparison: the same catch on the bare line (k ≈ 100 kip/ft) peaks at F = v·√(k·m) ≈ 26 kip. The limiter buys roughly a 5:1 reduction.
Takeaway: even with a good limiter the peak is ~4× static. Everything in the load path — line MBL, thimbles, shackles, padeye, winch and its mount, and the helical screw's holding capacity — must be sized for the limited peak (here ≈5.3 kip) with a proper safety factor, not for the 1.3-kip static tension. Snatch yanks are exactly what extracts poorly-set anchors; size embedment accordingly.
Alternatives & backups worth considering
| Option | Deadband (stiff in normal waves) | Dissipates energy | Self-resetting | Caps taut-line (trough) overloads | Complexity |
|---|---|---|---|---|---|
| Ball-detent preloaded spring (your concept) | Yes | Stores energy; heave plates + hysteresis dissipate it | Yes | Yes | Medium |
| Preloaded elastomer or Belleville stack | Yes | Yes (material hysteresis) | Yes | Yes | Medium-low |
| Slack standby snubber (two-stage line) | Yes — primary line carries normal loads | Partially | Yes | No | Low |
| Winch slip-drag + automatic re-tension | Yes (threshold in software/hardware) | Yes — best of all | Yes (motor rewinds) | Yes | Medium-high; needs power |
| Series nylon insert / ordinary snubber | No — soft all the time | Yes | Yes | Yes | Lowest |
| Sacrificial fuse (shear pin / breakaway) | Yes | Yes, once | No | Yes | Low — last-resort tier only |
A. Slack standby snubber (two-stage line) — highly recommended as the fail-safe
A second line a few inches longer than the primary, made of nylon or a rubber snubber with an HMPE backup strop. It triggers geometrically (the primary must be slack and the corner must rise Δ), so it can never false-trip in normal waves, it doubles as the fail-safe if the limiter fails, and it costs almost nothing. Two caveats: it does nothing for taut-line trough overloads; and keep Δ small (2–4 in), because after the elastic returns the platform, the primary re-catches from height Δ — small enough to be trivial.
B. Winch slip-drag ("fishing-reel drag") — you already own the hardware
Each corner has a motor unit between its pair of mooring screws. Give that winch a torque-limited slip mode — an adjustable friction clutch, or current-limited motor control — set just above the normal maximum tension. A snatch then pays line out at near-constant force, which is the ideal load limiter: it dissipates energy (no rebound, unlike a spring), the threshold is adjustable in software, and the motor simply re-tensions afterward. It also enables an operational trick: if you see a big wake coming, command the corners to slacken or drop the drag threshold before it arrives, ride it out free-floating, and re-tension after. Keep the mechanical ball-detent limiter as the no-power backstop — software and clutches fail, springs don't.
C. Series nylon insert / ordinary snubber
The standard yacht answer to snatch loads (a nylon snubber on an all-chain anchor rode). Simple and effective, but compliant at all loads — it fails your "no stretch in normal waves" requirement — and nylon heats under hard cyclic loading. Fine as the elastic element in the standby strop, wrong as the primary solution.
D. Preloaded elastomer or Belleville stack
Same physics as your coil spring in a smaller package. A stack of coned disc (Belleville) springs gives high force in short travel; stacked rubber/polyurethane pucks give a rising rate and free damping. Watch elastomer creep — re-check preload seasonally.
E. Sacrificial fuse — last tier only
A shear pin or breakaway link as the final protection against a beyond-design event is cheap insurance, but a blown fuse frees the corner — so always put it in series with the slack standby strop, never as the only protection.
F. Operational measures (the first line of defense)
- Tension for the worst expected condition: pull-down margin ≥ maximum expected crest. Your 3-ft policy makes the whole snatch problem rare; the limiter is the backstop for the misjudged or freak case.
- Line material: for permanently tensioned legs, prefer polyester over HMPE/Dyneema — HMPE creeps under sustained load and slowly loses your pre-tension. (The floating wind industry settled on polyester taut moorings for exactly these load-management reasons.)
- Monitor: a load cell per corner feeding your computer gives early warning of tension loss and a record of every limiter event.
Recommended arrangement (defense in depth)
- Policy: pull-down margin ≥ forecast maximum wave; re-tension as conditions change.
- Primary line: low-stretch polyester, sized for the limited peak with a healthy factor.
- Preloaded ball-detent limiter at each corner, above the waterline, co-located with the winch: threshold ≈ 1.3–1.5× normal max tension, ≥10–12 in travel, peak held to ~4× static, guided stem, polymer seat liner, housing with grease and drain, stroke tell-tale.
- Slack standby strop (Δ ≈ 3–4 in) rated for the full limited peak — fail-safe and helper in the catch.
- Winch slip-drag with automatic re-tension for energy dissipation and "wake coming" pre-slack — with the mechanical limiter as the no-power backstop.
- End-travel bumper inside the limiter, and a sacrificial fuse as the final tier.
This stack keeps the seastead genuinely rigid in sub-1-ft waves, caps any credible snatch at a load your hardware is actually designed for, dissipates the energy instead of bouncing it back, and fails safe if any single element lets go.
Commissioning tests
- Bench: cycle the limiter 100× at threshold with a come-along and a crane scale; verify trip force, smooth travel, reseating, and no galling. Re-test after a fouling season.
- Dock drop test: moored in calm water, slack one corner, hoist that corner 0.5–1 ft with a crane, quick-release, and record peak line load (load cell, or a phone accelerometer) with and without the limiter engaged. This directly validates the energy math above.
- Calibration: mark the preload screw with a reference scale so owners can set and verify the threshold; log every stroke event.