You tighten it. You mark it. Three months later the paint mark on the flange is a quarter turn off. If you have been chasing the same loose hardware on the same crane for years, the problem is not your torque wrench. It is that a bolted joint fails in two different ways, and the fix you reach for usually only addresses one of them.
A bolt is a spring, not a clamp
A bolted joint holds because the bolt is stretched. Tightening puts the fastener in tension, and that tension — preload — pulls the joined faces together hard enough that they behave like one part. The head is not gripping the surface. The bolt is a stiff spring under load, and everything the joint does depends on that spring staying stretched. Lose preload and the joint stops acting like one part. It starts acting like two parts with a pin through them.
Two failure paths, and only one involves rotation
Hardware gets loose two distinct ways, and confusing them is why so many fixes disappoint.
- Preload loss without rotation. The bolt never turns. Surface asperities on the mating faces flatten under load and embed. Gaskets and coatings creep. Thermal cycling and general relaxation take back a little more. The bolt is exactly where you left it, with less tension in it than you put there.
- Rotational self-loosening. The bolt actually turns out. This is the failure everyone pictures, and it is the one that ends with hardware on the deck.
Both end in a joint that will not hold. Only the second one backs a fastener out of its hole.
Why transverse vibration is the one that gets you
Axial vibration — load pulsing along the bolt axis — is comparatively benign. Transverse vibration, side to side across the joint interface, is the efficient loosening mechanism, and Junker testing is the standard method for demonstrating it.
The mechanism is straightforward. Under transverse load the joint faces slip against each other. Once that slip overcomes the friction holding the threads and the bearing surface, the fastener is briefly free to move. The thread is a helix, an inclined plane wrapped around a cylinder, and a preloaded bolt on an inclined plane wants to run downhill. Every slip cycle gives back a fraction of a degree. Thousands of cycles later you have a quarter turn, then a loose bolt, then nothing.
Cranes are built to deliver exactly that input. Travel vibration on runways, end trucks and trolleys. Shock loading through boom lacing and counterweight pins on every pick. Duty cycles in the thousands. Anything that survives that duty is surviving continuous transverse excitation, not an occasional bump.
What locking methods actually do
Prevailing-torque nuts, thread lockers, wedge lock washers and the drop risk they don't address, safety wire — these are real engineering, and good ones work. But note what they all target: rotation. They are vibration resistant fasteners in the sense that they resist the turning. That is the right goal for keeping a joint tight.
It is not the same goal as keeping a fastener out of the air. None of these methods addresses what happens after the joint has already let go — after embedment has eaten the preload, after a component was installed wrong, after the fastener finally worked out anyway. Safety wire compared with tethered bolts is the clearest illustration: wire restrains rotation, and a wire that has broken or been reinstalled poorly restrains nothing.
Retorque intervals have the same limit. They catch bolt loosening only when the interval is shorter than the failure. Between inspections, there is no control at all. Roughly 250 U.S. workers are killed by falling objects each year (U.S. Bureau of Labor Statistics), and most of that hardware passed its last inspection.
Keep the locking method. Add retention.
The mechanism argument leads somewhere specific: self-loosening is a joint integrity problem, and dropping is a consequence problem. Solve both. Keep whatever locking method your engineering calls for. Then, where a fastener sits over people, add secondary retention that does not care why the joint let go. A patent-pending engineered tether anchored to the structure keeps a backed-out fastener captive at height. It hangs where a technician can see it and recover it, instead of falling. Made to order in the USA, installed with standard tools and standard torque — see our approach to dropped object prevention for how that fits an existing joint design.
CraneBolts tethered safety bolts pair standard installation with patent-pending captive retention — made to order in the USA. Request a quote.