What a Turnbuckle and a Tensioner Are Doing in a Rig

Why this matters

A turnbuckle looks like a convenience. It is a length adjuster, and length adjustment is what you reach for when a load hangs crooked. What nobody tells a new rigger is that in any rig with more legs than the geometry needs, the adjuster is not levelling the load, it is deciding how the load splits between the legs, and a fraction of a turn moves a large fraction of a sling's rating from one leg to another. The rig that looks best hanging is frequently the one where two legs are carrying everything and two are along for the ride.

What a turnbuckle is rated for

A turnbuckle is a body with a right-hand thread at one end and a left-hand thread at the other, and two end fittings threaded into it. Its published working load limit is an in-line tension rating for the assembly as supplied, with both end fittings threaded to the manufacturer's required engagement.

Four things that rating assumes, each of which a rig routinely breaks.

The pull is in line. A turnbuckle jaw or eye pulled out of its own plane bends the same way any other fitting does, and the manufacturer's side-load reduction owns the factor. A sibling article covers that correction in full.

The end fittings are the ones it came with. Body and fittings are rated together. A jaw from one product line in a body from another is an unrated assembly.

Thread engagement is adequate. Most turnbuckle bodies carry an inspection hole through the body wall at each end. If you can see through it past the end of the threaded stud, engagement is short and the assembly is out of service. That check is the whole reason the hole exists, and it takes one look.

The end type is appropriate. Under ASME B30.26, Rigging Hardware, in whichever edition your jurisdiction, your contract or your employer's programme has adopted, jaw and eye ends are the ones that belong in a lifting rig. Hook-end turnbuckles are lower rated and the hook can come off the connection, and they are not lifting hardware.

Turnbuckles cannot be heated, welded, bent and straightened, or repaired, for the same reason no other forged rigging item can: all four alter the heat treatment or the section the rating was derived from.

Take-up is exact, and it is small

This is the one number in the subject you can derive rather than look up, and it explains the rest of the article.

Because one end is right-hand threaded and the other is left-hand threaded, a single full turn of the body draws each end fitting in by one thread pitch, so the assembly shortens by two times the pitch per full turn. A turnbuckle with a 1/4 inch pitch shortens by 1/2 inch per turn, and by 1/8 inch per quarter turn of the body.

Hold that against the other half of the relationship. The tension change in a leg is its axial stiffness multiplied by the change in its length, and the axial stiffness of a short steel sling leg is high. How high depends on the sling's construction, its diameter and its length, and the sling manufacturer's elongation data owns that figure, so this article gives you the relationship and routes the number. What you can bank without the number is the direction and the scale: on a short leg, an eighth of an inch of take-up is a large tension change, and on a long leg the same eighth of an inch is a small one. Adjusters are therefore most dangerous exactly where riggers reach for them most, which is on short legs in a tight headroom pick.

Determinate rigs and indeterminate ones

Two legs onto a rigid load is a determinate rig. The split between the legs is fixed by geometry, specifically by where the centre of gravity sits between the two pick points and by the leg angles, and no adjustment changes it. Shorten one leg and the load tilts; it does not transfer.

Three legs is the last configuration where geometry still governs, because three points define a plane. That is a statement about statics, not a sizing permission: on a rigid load the bridle sheet still says count two, because the geometry only governs while all three legs are genuinely in tension and you cannot confirm that by eye. Four legs onto a rigid load is one constraint more than the geometry needs, and the split is no longer set by geometry at all. It is set by the relative lengths of the legs, and leg lengths are never exactly equal: manufacturing tolerance, lug position tolerance, splice length, and any wear the slings have taken are all in the same order of magnitude as the take-up numbers above.

The trade's standing answer to that is conservative and it exists for exactly this reason: size a four-leg bridle on a rigid load as if two diagonal legs carry the whole load, unless the legs are individually adjustable and the distribution has been verified by measurement, by a qualified person, for that specific rig. Adjustability on its own does not qualify, because you cannot verify by eye that you tuned four legs into equal tension. That halves the number of legs you get to divide by, which doubles the share, which is the single most consequential sizing decision in a four-point pick.

The nose-down air handler

A long skid-mounted section came off a truck on a four-leg bridle to one hook. The angle convention is stated because half the trade quotes it from the vertical: all sling angles here are from HORIZONTAL. All four legs at 60 degrees from horizontal, load taken as 8,000 lb for the walkthrough.

What the crew sized it on. Four legs, so 8,000 / 4 = 2,000 lb of vertical share each. Leg tension is the share divided by the sine of the angle from horizontal, so 2,000 / sin 60 = 2,000 / 0.866 = 2,309, carried as 2,310 lb because rounding a tension upward is the conservative direction. Slings rated 3,200 lb each. Utilisation 2,310 / 3,200 = 72 percent. Comfortable, and wrong before the pick started.

What the rule above says to size it on. Two diagonal legs carrying, so 8,000 / 2 = 4,000 lb of vertical share each. Leg tension 4,000 / 0.866 = 4,619, carried as 4,620 lb. Utilisation 4,620 / 3,200 = 144 percent. The rig as assembled was over its slings before anybody touched a turnbuckle.

What the crew did next. It hung nose-down, because the centre of gravity was not at the middle of the skid. They put a turnbuckle in each of the two legs at the low end and took up three full turns on each. Thread pitch 1/4 inch, so the take-up was 2 x 1/4 = 1/2 inch per turn, three turns, 1.5 inches shorter on each of those two legs.

What that did. The skid came level, and the two legs at the other end went visibly slack. Nothing about the load changed. The two shortened legs took up length until they were carrying essentially all of it, and the rig arrived at exactly the two-leg condition the sizing rule warns about, except now it was two legs at one end of a load whose weight was never evenly distributed to begin with. The turnbuckles did not level the load in the sense of balancing it. They levelled its attitude and concentrated its weight.

The turnbuckles themselves. Take those as rated 3,500 lb for the walkthrough, chosen to match the slings the crew thought they were sizing. Actual tension 4,620 lb. Utilisation 4,620 / 3,500 = 132 percent. A turnbuckle overloaded in tension distorts the body or strips the thread engagement, and a stripped engagement releases the leg completely and at once. There is no partial version of that failure.

The finding, and it was visible. A slack leg in a multi-leg bridle is the rig telling you what the split is. It is not a cosmetic problem to be tidied up with the adjuster; it is a measurement. When two legs of four go slack, the honest reading is that the two remaining legs are carrying the whole load, and the correct response is to land the load and re-plan rather than to adjust until it looks right.

What the correct sequence would have been. Land the load and block it. Establish where the centre of gravity actually is rather than assuming the middle of the skid, which a sibling HowTo covers as its own procedure. Move the pick points or set the leg lengths with the load down so the geometry does the work, then take a trial lift to a small clearance with everybody outside the swing arc and confirm all four legs come into tension. Size the slings for the two-leg case anyway, because a verified distribution on the ground is not a guarantee once the load flexes.

The safety finding, which is the one that would have hurt somebody. The crew turned those turnbuckles with the load suspended. There is no way to reach an adjuster in a hanging rig without standing under a suspended load, which for construction cranes and derricks is the duty at 29 CFR 1926.1425 and in general industry sits in the load-handling requirements at 29 CFR 1910.179 for overhead and gantry cranes. Adjustment happens with the load landed and blocked and every leg slack, never at height and never on a live rig. A hoist brake holding the load while somebody works underneath is not a parking device, and a load left hanging while the rigging is fiddled with is the specific condition those rules exist to prevent.

Locking, and the failure that takes an hour

A turnbuckle in tension does not spontaneously unwind, but a turnbuckle in a rig that rotates does. Any load free to spin about its hook winds and unwinds the rigging below it, and an unlocked turnbuckle in one of those legs slowly lengthens or shortens while nobody is watching. On a long hold, that is enough to shift the distribution in a four-leg rig or to run a body out past its thread engagement.

So every turnbuckle in a lifting rig gets secured against rotation, by the manufacturer's locking method: jam nuts run up against each end of the body, or safety wire through the body and around both studs, or a locking device supplied with the product. Wire the body to the studs rather than the studs to each other, because the studs are what has to be prevented from turning relative to the body.

Rotation itself is worth understanding rather than just resisting, and a sibling article covers where it comes from and what a tag line does and does not control.

The same device, a different job

Everything above is about a turnbuckle inside a load path that is being lifted. A turnbuckle used as a tensioner in guying, bracing, or a hold-down is doing a genuinely different job: there the whole point is to establish and maintain a preload, adjustment under tension is the function rather than a violation, and the target tension usually comes from whoever designed the guy or the brace.

Two things do not change across that fork. The rating is still an in-line tension rating and side load still reduces it. And the locking requirement is still there, because a guy that slowly slackens is a guy that has stopped doing its job, and nobody notices until the thing it was bracing moves.

Where a bracing or guying tension is specified, that specification belongs to a registered professional engineer or to the equipment manufacturer, and the rigger's job is to reach it and lock it rather than to choose it.

References

• 29 CFR 1926.1425, Keeping clear of the load (construction cranes and derricks), and 29 CFR 1910.179 for general industry overhead and gantry cranes
• 29 CFR 1926.251, Rigging equipment for material handling (construction), and 29 CFR 1910.184, Slings (general industry): inspection before use and safe operating practice for the slings the adjuster sits in
• ASME B30.26, Rigging Hardware, in the edition adopted by your jurisdiction, your contract or your employer's programme: turnbuckle identification, end types, engagement and removal criteria
• Turnbuckle and sling manufacturer's catalogue: working load limit by size and end type, required thread engagement, locking method, and sling elongation data
• See related: How to Find a Load Center of Gravity Without a Drawing; Why a Load Rotates and What a Tag Line Actually Controls