Self-Loosening of Threaded Fasteners

In this article, Bill Eccles of Bolt Science examines why threaded fasteners come loose in service — and what engineers can do about it at the design stage. The article distinguishes between two fundamentally different loosening mechanisms and explains why confusing them leads to ineffective countermeasures.

Threaded fasteners have been self-loosening since the start of the industrial revolution, and engineers have been devising locking methods for just as long. Many of those methods are still widely used today — and many of them don’t work as well as commonly assumed. Understanding the actual mechanics behind loosening is the prerequisite for specifying a solution that will hold.

What the article covers:

• Rotational vs. non-rotational loosening — These are distinct failure modes with different causes. Non-rotational loosening involves no thread movement; it results from embedding — the partial plastic collapse of surface asperities under bearing load. This typically accounts for 1–5% of clamp force loss within seconds of tightening. Rotational loosening, by contrast, is driven by transverse joint movement that overcomes the frictional resistance generated by bolt preload, allowing the fastener to back out under cyclic loading.
• Junker’s foundational research — In 1969, Gerhard Junker published work demonstrating that transverse dynamic loads are far more damaging to bolted joints than axial loads. His key finding: self-loosening begins the moment relative motion occurs between the mating threads and bearing surfaces. The Junker test machine — still the basis of DIN 65151 testing — quantifies loosening resistance by applying controlled transverse displacement while continuously monitoring preload loss.
• Why spring washers often make things worse — Junker test data shows that helical spring washers can actually accelerate loosening compared to an unmodified bolt. Many major OEMs have removed them from internal standards as a result, though their use remains widespread in the broader industry.
• The real solution: adequate preload — A joint designed so that bolt preload generates enough friction grip to prevent transverse movement will not self-loosen, and no locking device is needed. The article walks through how friction variation, tightening torque tolerances, and prevailing torque interact to define the minimum and maximum preload range achievable from a given assembly specification — and why designing to the minimum anticipated preload is the only way to eliminate loosening risk entirely.
• When locking devices are warranted — In applications where joint movement cannot be prevented (thermal expansion being a common example), a locking device of proven performance should be specified. The article provides design guidance for making that selection on technical rather than conventional grounds.

These principles apply wherever bolted joints are used under dynamic loading — from transportation and heavy equipment to structural steel and industrial machinery. The mechanisms don’t change with the application; only the consequences of overlooking them do.

About Bolt Science 

Bolt Science was founded in 1992 to provide independent technical expertise in bolted joint technology. The company offers bolted joint analysis software, consulting and problem-solving services, fastener and joint testing, and training on bolting technology. Their client list includes many of the world’s major engineering organizations. Learn more at boltscience.com.