01
What actually happens during an upend
A vessel rotates from horizontal to vertical while two cranes share it, and the share moves continuously from one to the other. The procedure below is the same on every upend; what changes is where each check peaks.
The operation, in the order it happens:
Flat, both cranes taking load. The vessel lies horizontal on stands or a transport frame. The main crane is on the top attachment, the tail crane or the tailing device is at the bottom. The share between them follows from where the two attachments sit relative to the centre of gravity, in exactly the way a simply supported beam's reactions do.
The tail lifts and the rotation starts. The vessel begins to pivot about the tail. The main hook takes an increasing share, the tail a decreasing one, and both hooks have to travel horizontally to stay over their attachments.
Through the middle. The sling to the tailing attachment is now far out of the plane of that attachment, because the vessel has rotated underneath it. This is where the tailing lug's worst case usually sits.
Approaching vertical. The main hook is carrying nearly everything. The tail crane is nearly unloaded and the tail attachment may be released.
Vertical, transferred. The main crane has the whole vessel.
Two things about that sequence make it different from an ordinary lift.
Nothing is constant. The angle of every sling, the share on every hook, the stress in the shell and the direction of load on the tailing attachment all vary continuously with rotation.
The cranes have to move. Both hooks travel horizontally, which means both cranes are changing radius, which means both are moving on their own load charts through the operation.
02
Angle by angle: why there is no single worst case
Four checks, four different governing angles, spread across the whole rotation. Any calculation performed at one inclination is a calculation of something other than the operation.
Vessel Upending & Tailing Lift Calculator · computed at page render
A 60 t vessel, two-crane upend, swept
The calculator solves the whole rotation and reports each check at the angle where it is worst.
| Main hook loadworst at 90 degrees, at the top of the rotation | 600.0kN |
|---|---|
| Main hook utilisation | 80.0% |
| Tail hook loadworst at 0 degrees, flat on the stands | 252.2kN |
| Tail hook utilisation | 63.1% |
| Tailing lug forceworst at 70.4 degrees out of the lug's plane | 224.7kN |
| Tailing lug utilisation | 75.9% |
| Body bending stressworst at 6.5 degrees, barely off horizontal | 15.8MPa |
| Sling tension against declared WLL | 75.0% |
Read the four governing angles: 90, 0, 70 and 6.5 degrees. They are spread across the entire rotation, and no snapshot catches more than one of them.
Open this example in the calculatorThe practical rule: sweep it. If a calculation for an upend does not say what angle each governing value came from, it is not a calculation of an upend. That is true whether the sweep is done by software or by hand at a sensible interval; what is not acceptable is three snapshots and a hope.
03
Why the tailing lug is the hardest component
Because through most of the rotation the sling is pulling it far out of its own plane, and a lug loaded out of plane is a lug being asked to do the thing it is worst at.
A padeye is efficient because it is a plate, and a plate is efficient in its own plane. Out of plane it is a cantilever with a small section modulus, and every degree away from the plane trades a good load path for a bad one.
On the worked upend the tailing lug's worst case is at 70.4 degrees out of plane, carrying 224.7 kN. That is not an unlucky instant; it is a substantial fraction of the rotation.
Three consequences that explain what tailing lugs look like:
They are thick. Out-of-plane bending needs section, and section on a plate means thickness.
Their welds are large and returned. The moment goes into the weld group through the lever arm, and out-of-plane loading puts the peak stress at the weld ends.
They are often not lugs at all. A trunnion is the natural answer to a sling that has to change direction through tens of degrees, which is exactly what a tailing attachment does. That choice belongs at the start of the design, not after the lug has failed.
Choosing between a trunnion and a lug turns on exactly that question, and an upend is rotation by definition.
04
The one real design choice
Where the top attachment sits. Everything else about an upend is a consequence, and this is the input that moves the answer most.
Vessel Upending & Tailing Lift Calculator · computed at page render
The same vessel with the top attachment 3 m higher
Identical vessel, identical cranes, identical tail arrangement. Only the top attachment's position along the shell changed.
| Top attachment positionagainst 25.0 m from the tail | 28.0m |
|---|---|
| Tail hook loadagainst 252.2 kN | 290.2kN |
| Tail hook utilisationagainst 63.1% | 72.5% |
| Tailing lug forceagainst 224.7 kN | 260.4kN |
| Tailing lug utilisationagainst 75.9% - now governing | 88.2% |
| Main hook utilisationunchanged: the main hook still takes the whole vessel at the top | 80.0% |
Three metres moved the tailing lug from 76 percent to 88 and handed it the governing position. The main hook did not move at all, because at the top of the rotation it carries the whole vessel regardless of where the attachment is.
Open this example in the calculatorThat asymmetry is worth holding onto:
The main hook's worst case is fixed. At the top of the rotation it carries the vessel, and no attachment position changes that.
Everything else is a lever. The tail hook, the tailing lug and the body stresses all respond to where the top attachment sits, and they all move in the same direction.
So the position is chosen against the tail, not the main. Moving the top attachment towards the tail unloads the tail crane and the tailing lug, at the cost of a longer cantilever of vessel above it, which shows up in body bending.
05
The tail, and the number nobody knows
Tail resistance is a band. Declaring one value is claiming a measurement nobody took.
Whether the tail is dragged on skids, rolled on a dolly, or lifted by a second crane, the horizontal resistance at the tail affects the whole force balance. And unlike the geometry, it is genuinely unknown:
- Skids on steel and skids on timber are different materials in contact, not the same arrangement with a different finish.
- A dolly with good bearings and a dolly that has stood in a yard through a winter can differ by more than the difference between skidding and rolling.
- Wet, iced or contaminated surfaces are a different problem again.
- The tail digging in as the vessel rotates is a different mechanism from sliding.
The honest treatment is a declared band, and a requirement that the operation is acceptable at every value in it. A single value is a claim that somebody measured the friction, and nobody did.
A wide band is a better answer than a precise wrong one. If the sweep passes across a resistance range of 0.2 to 0.7, the operation is robust. If it only passes at 0.35, the design depends on a number nobody can supply.
06
What an upend needs in its plan
Everything a critical lift needs, plus seven things that only exist because the geometry moves.
Additional to a normal critical lift plan
- 01The sweep, not a snapshotEvery check reported with the angle it governs at, across the whole rotation.
- 02Both cranes' duties through the rotationBoth hooks travel horizontally, so both cranes change radius. The governing duty is somewhere in the middle, not at either end.
- 03The tail resistance bandA declared range with the operation acceptable at every value in it, not a single assumed number.
- 04The out-of-plane angle at the tailing attachmentStated as a maximum, checked at that maximum, and on the drawing where a rigger can see it.
- 05The transfer pointThe angle at which the tail is released, who confirms it, and what happens if it is released early.
- 06The body's own stressesA vessel is a thin shell being bent about its own axis. The governing case is near horizontal, not near vertical.
- 07Clearances through the arcThe vessel sweeps a large area. The worst clearance is rarely at the start or the end.
07
Seven ways an upend goes wrong
Most of them are one error wearing different clothes: treating a continuous operation as a small number of positions.
1. Three snapshots instead of a sweep. Start, middle and end miss at least one of the four governing angles in the worked example.
2. The tailing lug checked in plane. Its worst case is tens of degrees out of plane, which is the load case it is worst at.
3. One tail resistance value assumed. It is a band, and nobody measured it.
4. The cranes' duties checked at one radius. Both hooks travel through the rotation, so both machines move on their charts.
5. Body stresses checked vertical. A vessel's bending is worst near horizontal, which is the position that looks least dramatic.
6. The transfer point never stated. The moment the tail is released is a discrete event with its own load case, and it needs a stated angle and a person who confirms it.
7. The top attachment position treated as fixed. It is the one genuine design lever, and moving it 3 m changed the governing check in the worked example.
Common questions
- At what angle is a vessel upend worst?
- There is no single angle, which is the whole difficulty. On the worked 60 t upend the main hook load peaks at 90 degrees at the top of the rotation, the tail hook at 0 degrees flat on the stands, the tailing lug at about 70 degrees out of its own plane in the middle, and body bending at 6.5 degrees barely off horizontal. Those four are spread across the entire rotation, so any calculation performed at one inclination is answering a different question from the one the operation asks.
- Why are tailing lugs so heavy?
- Because through most of the rotation the sling pulls them far out of their own plane, and a plate loaded out of plane is a cantilever with a small section modulus. On the worked upend the tailing lug's governing case is about 70 degrees out of plane. That is why they are thick, why their welds are large and returned, and why a trunnion is often the better answer for a tailing attachment - rotation is the deciding question, and an upend is rotation by definition.
- How do I decide where to put the top attachment on an upend?
- It is the one genuine design choice in an upend and it is chosen against the tail rather than the main. Moving it towards the tail unloads the tail crane and the tailing lug, at the cost of a longer cantilever of vessel above it, which shows up in body bending. In the worked example moving it 3 m further from the tail took the tailing lug from 76 to 88 percent and the tail hook from 63 to 73, while the main hook did not move at all because at the top of the rotation it carries the whole vessel regardless.
- What value should I use for tail friction?
- A band, not a value, and the operation should be acceptable across all of it. Skids on steel and skids on timber differ substantially, a dolly with good bearings and one that has stood in a yard differ substantially, and wet or iced surfaces are different again. A design that only passes at one precise resistance value depends on a number nobody measured; a design that passes across a wide declared band is robust.
- What extra does an upend need in its lift plan?
- Beyond a normal critical lift plan: the sweep rather than a snapshot, with each check reported at the angle it governs at; both cranes' duties through the rotation, because both hooks travel horizontally and both machines change radius; the declared tail resistance band; the maximum out-of-plane angle at the tailing attachment, stated on the drawing; the transfer point at which the tail is released and who confirms it; the vessel's own bending, which is worst near horizontal; and clearances through the whole arc.
Sources
Every document below is linked at its publisher or regulator. Xarpis reproduces no standard text; where a clause is named, the identifier is given so you can find it in your own copy.
DNV-ST-N001Marine operations and marine warranty
DNV · paid document
The marine warranty standard behind most offshore lift factor sets: dynamic amplification, skew load and consequence factors, and the load cases a marine operation is planned against. Widely applied onshore by contract even though its scope is marine.
HSE lifting equipment guidanceLifting equipment at work: planning and organising lifting operations
UK Health and Safety Executive · free to read
The regulator's own plain-language account of what planning a lifting operation means in UK law: who is competent to plan it, what a plan has to address, and how it scales from a routine repetitive lift to a one-off complex one. Free, short, and the closest thing to an official answer to 'what has to be in a lift plan'.
WRC 537Precision Equations and Enhanced Diagrams for Local Stresses in Spherical and Cylindrical Shells Due to External Loadings
Welding Research Council · paid document
The current basis for local stresses in a cylindrical or spherical shell from an external attachment loading: the calculation a trunnion welded to a vessel needs, and one that no pressure-vessel construction code supplies directly. It supersedes the presentation in WRC 107 rather than the physics, and the pressure-vessel codes reference it as an accepted local-stress method rather than reproducing it.
ASME B30.5Mobile and Locomotive Cranes
ASME · paid document
Construction, installation, operation, inspection and maintenance of mobile cranes in the US, including load rating and the requirement to operate within the manufacturer's chart. It governs the machine; the ground it stands on is 29 CFR 1926.1402 and the calculation is yours.
Run the check properly
Reading about a calculation is not the same as being able to hand one over. These tools produce the traceable record.
Something here wrong, or thinner than it should be? Tell us which paragraph and it gets rewritten. Articles carry the date they were last revised for exactly this reason.