01
Why an eccentric load tilts, and what it stops at
A suspended body has one equilibrium: centre of gravity directly under the hook. Everything else is a description of how far it has to rotate to get there.
Hang anything from a single point and it settles with its centre of gravity vertically below that point. There is no other stable position, and the load does not need anyone's permission to find it. The assembly simply rotates until the horizontal offset between the hook and the combined centre of gravity is zero.
The equilibrium the load is looking for
- is the tilt angle the assembly settles at
- is the combined centre of gravity of beam, slings and load, which moves as it tilts
The right-hand side depends on the answer, which is why a tilt solve is an equilibrium problem rather than a division.
For a spreader beam, that means the whole assembly - beam, slings and load - rotates until the combined centre of gravity is under the hook. The tilt is not a defect in the rigging; it is the rigging's answer to being asked to hold an off-centre load from a centred hook.
Three things change when it tilts, and they compound:
The sling angles change. One leg gets steeper and the other flatter, because the hook has moved horizontally relative to the lift points.
The tension split gets worse. The flatter leg carries more, on top of already carrying the larger share.
Whatever is being lifted is now at an angle. For a set-down onto a level foundation, or an item that must not be tilted, that is the whole problem.
02
What a level solve tells you, and what it assumes
It answers what the legs carry if the arrangement is held level. That is a real question, and it is only the right one if something is actually holding it.
Take the calculator's 10 t, 6 m spreader and move the centre of gravity along the beam, keeping the arrangement level.
Spreader Beam Design Calculator · computed at page render
Centred, solved level
The reference. Symmetric, so level costs nothing to hold.
| Tension, left top sling | 60.1kN |
|---|---|
| Tension, right top sling | 60.1kN |
| Sling angle, left | 59.3deg |
| Sling angle, right | 59.3deg |
| Peak moment in the beam | 8.2kN·m |
Spreader Beam Design Calculator · computed at page render
900 mm off centre, still solved level
Fifteen percent of the span. The solve holds the beam level, which is a condition somebody would have to impose.
| Tension, left top slingagainst 60.1 kN centred | 46.4kN |
|---|---|
| Tension, right top sling | 72.2kN |
| Sling angle, leftflatter | 52.6deg |
| Sling angle, rightsteeper | 67.0deg |
| Sling angle limitagainst 75.9% centred | 85.5% |
Everything passes and nothing on this page says that holding it level requires anything. That is the assumption worth being explicit about, because it is doing a lot of work.
Open this example in the calculatorA level solve is a statement about a constraint. Something has to supply it: a second hook, a tirfor, an adjustable leg set to a computed length, or a top lug moved along the beam so the hook lands over the centre of gravity. If nothing does, the level solve is describing a condition that will not exist.
03
What the arrangement actually does
Release the same case and it rotates. In the worked example, 8.2 degrees, which fails the arrangement's own tilt limit.
Spreader Beam Design Calculator · computed at page render
The same 900 mm offset, released
Identical geometry, identical load, identical slings. The only change is that the solve is allowed to find the equilibrium tilt instead of being told the beam is level.
| Equilibrium tiltfound by the solve, not assumed | 8.19deg |
|---|---|
| Tension, left top slingagainst 46.4 kN held level | 44.4kN |
| Tension, right top slingagainst 72.2 kN held level | 73.8kN |
| Sling angle, left | 51.8deg |
| Sling angle, right | 68.2deg |
| Peak moment in the beamagainst 7.8 kN·m held level | 8.6kN·m |
| Equilibrium tilt checkagainst the arrangement's declared tilt limit | 136.4% |
The check that fails is the tilt itself, not the steel. That is the correct output: the arrangement is not overstressed, it is unfit for a lift that needs the load to arrive level, and it says so.
Open this example in the calculatorNotice which direction the tilt pushed the tensions. Held level the split was 46.4 against 72.2 kN; released it is 44.4 against 73.8. The tilt made it worse. The rotation steepens the already-loaded leg and flattens the other, and the flattening is what costs.
04
Five ways to fix it, cheapest first
Four of them are rigging changes and one is structural. The rigging changes remove the tilt; the structural one only survives it.
Move the top lug. On a spreader with an adjustable or multi-hole top connection, sliding the hook attachment along the beam until it sits over the combined centre of gravity removes the tilt entirely. This is the cleanest fix and it needs no extra hardware.
Adjust a sling length. Shortening the leg on the light side and lengthening the other rotates the beam back to level. Chain shorteners and turnbuckles make this a field adjustment, but it needs a computed target rather than a guess, and the adjustment must be recorded.
Move a lift point on the load. Where the load's attachment points are yours to place, placing them symmetrically about the centre of gravity solves the problem before it exists. It is the earliest and cheapest intervention and the one most often missed, because the centre of gravity is usually calculated after the lift points are drawn.
Accept the tilt and design for it. Legitimate when the load does not care, and it means checking every check at the tilted geometry: sling angles, beam actions, clearances and the set-down. It also means the tilt goes on the plan as an expected condition rather than a surprise.
Use a second hook or a tailing crane. Two suspension points constrain the rotation. This is a different operation with its own load sharing problem, and it is the answer when the load genuinely must stay level.
Before signing off an eccentric pick
- 01State the centre of gravity and its toleranceA position with an envelope, and the worst position in that envelope is the one checked.
- 02Say whether the solve was level or freeIf level, name what supplies the constraint. If free, report the tilt.
- 03Check the tilted geometry, not the nominal oneSling angles, beam actions and clearances all recomputed at the tilt the arrangement settles at.
- 04Check the flatter legThe rotation flattens one leg, and the flatter leg is where the angle factor bites.
- 05Check the set-downA tilted load meeting a level foundation lands on one corner. Say how it is brought level, and when.
- 06Record any sling adjustment as a valueA computed length with a tolerance, not an instruction to adjust until it looks right.
- 07Check clearances at the tiltA tilted load is longer in plan and lower at one end than the drawing shows.
05
What the rules expect you to have thought about
Not a tilt limit. They expect the load's properties to be established and the operation planned by somebody competent, and an eccentric pick is exactly the case those words were written for.
The declaration is the point. A spreader beam that will tilt has to have been designed knowing that, and a lift plan for an eccentric pick has to say which of the five fixes above is being used. "It will hang slightly off level" is not a plan; it is a prediction with nobody's name on it.
06
Six ways an eccentric pick goes wrong
Five of them are the same mistake: solving a condition rather than the operation.
1. A level solve with nothing holding it level. The commonest, and it is invisible on the output unless somebody asks what supplies the constraint.
2. The tilt calculated and the geometry not rechecked. Knowing the beam sits at 8 degrees is worth nothing until the sling angles, the beam actions and the clearances are recomputed there.
3. The CoG taken as a point. A centre of gravity has a tolerance, and the tilt at the edge of that tolerance is the one to design for. A cog offset stated without an envelope cannot produce a governing case.
4. The flatter leg overlooked. The rotation steepens one leg and flattens the other, and it is the flat one whose angle factor climbs.
5. Sling adjustment left to the field. "Adjust until level" produces an unrecorded geometry, and the calculation then describes a different arrangement from the one in the air.
6. The set-down not thought about. A tilted load meeting a level foundation lands on one corner, which is a point load nobody calculated on a structure nobody warned.
Common questions
- Why does an off-centre load hang at an angle?
- Because a suspended body has exactly one equilibrium: its centre of gravity directly below the hook. If the rigging does not put the hook there, the assembly rotates until it does. The tilt is not a defect in the rigging, it is the rigging's answer to being asked to hold an off-centre load from a centred hook, and it changes the sling angles, the tension split and the attitude of the load all at once.
- What is the difference between a level solve and a tilt solve?
- A level solve answers what the legs carry if the arrangement is held level, which is a real question only if something is actually holding it - a second hook, an adjustable leg set to a computed length, or a top lug moved over the centre of gravity. A tilt solve answers what the arrangement will do when it is released. On the worked 900 mm offset the level solve reports 46.4 and 72.2 kN, and the tilt solve reports 44.4 and 73.8 kN at 8.2 degrees of tilt.
- Does tilting make the sling tensions better or worse?
- Worse. The rotation steepens the already more heavily loaded leg and flattens the other, and it is the flattening that costs, because a flatter leg has a larger angle factor. In the worked case the split went from 46.4 and 72.2 kN held level to 44.4 and 73.8 kN released, so the difference between the legs widened rather than evened out.
- How do I stop an eccentric load from tilting?
- Five options, cheapest first. Move the top lug along the beam until the hook sits over the combined centre of gravity, which removes the tilt entirely and needs no extra hardware. Adjust a sling length to a computed target and record it. Move a lift point on the load, which is the earliest and cheapest intervention. Accept the tilt and check every check at the tilted geometry. Or use a second hook, which constrains the rotation and introduces its own load sharing problem.
- What should a lift plan say about an eccentric pick?
- Which of the fixes is being used, and what supplies it. It should state the centre of gravity with its tolerance and confirm the worst position in that tolerance was checked, say whether the solve was level or free, report the tilt if free, and record any sling adjustment as a computed length with a tolerance rather than an instruction to adjust until it looks right. It should also say how a tilted load is brought level for set-down, because a tilted load meeting a level foundation lands on one corner.
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.
LOLER 1998Lifting Operations and Lifting Equipment Regulations
UK Health and Safety Executive · free to read
The UK duty framework for lifting operations: planning by a competent person, supervision, and thorough examination of lifting equipment and accessories. Like OSHA's rules it governs the process, not the arithmetic.
EN 13155Cranes. Safety. Non-fixed load lifting attachments
BSI (national adoption of the CEN standard) · paid document
The harmonised European standard for non-fixed load lifting attachments - the family a spreader beam or lifting beam belongs to. It carries the load basis and the proof requirements, not member resistances, which is why an EN route needs EN 1993 alongside it. Now published as EN 13155:2020+A1:2025.
ASME BTH-1Design of Below-the-Hook Lifting Devices
ASME · paid document
Structural, mechanical and electrical design criteria for below-the-hook lifting devices, used alongside ASME B30.20 which carries the safety requirements. The current edition is BTH-1-2023; Xarpis implements the 2020 edition and says so on every result.
ASME B30.9Slings
ASME · paid document
The US volume covering alloy steel chain, wire rope, metal mesh, synthetic rope, synthetic webbing and synthetic round slings: rated loads, marking, inspection, and the removal criteria that decide when a sling leaves service. Where published sling rated loads and angle reductions come from.
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.