What a Tandem Lift Adds Beyond Two Cranes

Why this matters

Two cranes on one load are not two lifts happening beside each other. The load couples them, so the share on each hook is set by geometry rather than by agreement, and every time the geometry changes the share moves. That is why the instant to check is almost never the pick, and why a crew can plan carefully, size both machines with room to spare, and still put one of them into alarm partway through the operation. What follows is one job that did exactly that, what the numbers said afterwards, and the two coupling mechanisms it exposed.

The signal

A fabrication shop was upending a 40-foot vessel from horizontal to vertical on its foundation. Main crane on the top-end lug, a smaller tailing crane on a bottom-end lug, both on outriggers, both with load indication. The plan had been checked at the pick and both machines had comfortable margin.

At roughly 60 degrees of rotation the main machine's load-moment indicator went into alarm. The lift director stopped everything, held the piece, and the crew spent the next hour bringing it back down to horizontal under controlled load lowering, with the exclusion zone extended and nobody permitted inside the radius of either machine or in the arc the top end would sweep if it got away. Nothing broke. The lift went the following week after a re-plan.

The geometry the plan checked

The vessel weighed 18,000 lb. Picked horizontal, the two hooks share it by the position of the centre of gravity between the pick points.

  base end top end
     |                                     |
   tail pick (2 ft)          main pick (38 ft)
     |                                     |
  ===+=================o===================+===
                       |
              CG, 16 ft from base end

     |<-------- 36 ft pick to pick -------->|

• Span between picks: 38 minus 2 = 36 ft
• Tail is 14 ft from the CG, main is 22 ft from the CG
• Share on the tail hook = 18,000 x 22 / 36 = 11,000 lb
• Share on the main hook = 18,000 x 14 / 36 = 7,000 lb
• The two shares sum to 18,000 lb, which is the only check on the arithmetic that matters

Note which machine carries more. The hook nearer the centre of gravity takes the larger share, so the smaller tailing machine was doing most of the work at the pick. That is normal and it is the reason a tailing crane is not automatically the small one.

What the plan did not check

At the end of a clean upend, the vessel hangs vertical on the main hook and the tail hook is carrying nothing. The whole 18,000 lb has transferred to the main machine.

• Main hook at the pick: 7,000 lb
• Main hook at the finish: 18,000 lb
• Ratio through the operation: 18,000 / 7,000 = 2.57 times

The crew had sized the main machine against 7,000 lb plus rigging. Their margin was real and it was against the wrong instant. The governing check on a tandem lift is the worst combination of hook load and radius reached at any point in the operation, and on an upend that is at the end, when the load is fully on one machine and the boom is usually out further than it was at the pick.

The two ways the machines are coupled

The share migration above is the arithmetic. Two mechanisms make it worse in the field, and both are operator inputs rather than plan errors.

Unequal hoist speed transfers load. If the main hoists faster than the tail pays out, the piece rotates faster than the plan assumed and the tail line goes slack. A slack line that comes back into tension is a shock load, and a rated capacity is a static rating that already contains its design factor computed for a static pull. Shock defeats it, and 29 CFR 1910.184(c) states plainly that slings shall not be shock loaded. The construction counterpart for rigging equipment is 29 CFR 1926.251. Neither Part gives you a shock allowance to spend, because there is no way to bound the peak.

The tail pick point travels horizontally. As the piece rotates upright, the bottom lug moves toward the main crane. The tail machine must keep its load line plumb by travelling, booming in, or swinging, under the lift director's direction and only by a method the machine's manufacturer permits with a suspended load. If it does not, the line goes out of plumb and starts dragging the piece sideways, which side loads that boom. Load charts are computed for a load hanging in line with the boom point, so a side-loaded boom is off the published condition with no published answer for where it now is.

The derate, and the order the corrections go in

A multiple-crane lift is planned with each machine derated below its chart figure, because the share on either hook is less certain than a single-crane hook load and because a small geometric error moves load between the machines rather than just adding to one. A derate to 75 percent of the machine's available capacity is a common planning value. It is not a rule you own: it belongs to the lift plan and the qualified person who signs it, and where a manufacturer publishes its own multiple-crane guidance, that governs instead.

Run it on the main machine at the finishing radius, with every correction on its own line.

Two things about that table are deliberate and both run conservative. The deductions come off before the derate, so the derate applies to what is actually available rather than to a gross figure that still has a hook block inside it. And the chart figure is a re-basing, not a starting point: a load chart already contains the machine's own weight and structure but never contains anything hanging below the boom point, so block, ball, slings, spreaders and a stowed jib all come off it. That correction is a subtraction from the machine's side, not an addition to the load's side, and a sibling card on published weights covers why the direction matters.

Even with the derate applied correctly, 6.8 percent is a thin margin for a lift whose hook load moves by a factor of 2.57 during the operation. The re-plan moved the main machine in to a shorter radius rather than accepting it.

The same migration reaches the ground. What a mobile crane puts through its outriggers is its own weight plus the hook load, distributed by where the boom is pointing, so the main machine's ground loading rises through the upend by the same 11,000 lb the hook gained while the tail machine's falls away. A ground bearing assessment made at the pick has assessed the easiest moment for one machine and the hardest for the other. Mat sizing and allowable bearing pressure belong to a registered professional engineer working from a real subsurface assessment, and the question to hand that engineer is not "what do these two cranes weigh" but "what does each one carry at each stage of the rotation." Where a slab, a basement, a vault or a buried utility is under either machine, that is a structural question about the slab as well, and it is answered before the mats are laid rather than after a corner drops.

What made the difference on the second attempt

Three changes, and only one of them was arithmetic.

The plan carried the hook load at five stated points through the rotation rather than at the pick, so the crew could see the main machine's number climbing and the tail's falling before anything moved. The governing instant was printed on the plan next to the machine that owned it.

Both machines reported indicated load to the lift director on a dedicated channel at each of those points, and the operation paused at each one. Load indication on both machines is what turns a share calculation into something observable; without it, a transfer between the two hooks is invisible until one machine complains.

And the lift director had one voice and one signal person. On a tandem lift there is no such thing as a local decision: an operator who takes up a little slack on his own has moved load onto the other machine, and the other operator finds out from his indicator. Everyone stays clear of both machines' swing paths, which on a mobile crane includes the full arc of the counterweight tail swing as well as the load radius, and the pinch between a swinging counterweight and any fixed object is a fatal crush point that has nothing to do with the load at all.

What would have caught it the first time

The plan was correct at every point it examined. What it lacked was a reason to examine any point other than the pick, and that is the portable lesson: on any operation where the load changes orientation, ask which hook the weight ends up on and check that instant first, because a rotation is a load transfer with a slow fuse.

The same question has a different answer on a straight tandem carry, where both hooks hold their share the whole way and the governing instant really is the pick, provided neither machine's radius changes. Ask which one you have before you decide where to look.

References

• 29 CFR 1926 Subpart CC, cranes and derricks in construction, for operation in accordance with manufacturer procedures and for the qualified-person and signal-person requirements
• 29 CFR 1910.184(c) and 29 CFR 1926.251, slings, including the prohibition on shock loading
• ASME B30.5 for mobile cranes, in the edition adopted by your jurisdiction, your contract or your employer's programme, for multiple-crane lift direction
• The machines' own load charts and any manufacturer guidance on multiple-crane operation, which governs over any generic derate
• See related: Why a Published Weight and an Actual Weight Diverge; What Wind Does to a Lift and When It Stops It