Many state Departments of Transportation (DOT) across the US, including MnDOT, are experiencing problems associated with loose anchor bolts used in support structures (e.g., overhead signs, high-mast light tower (HMLT), and tall traffic signals). Specifically, MnDOT inspection crews have found loose nuts at most anchor bolt locations, even at some newly installed signs. Many of these nuts became loose in less than two years, even after being tightened by the maintenance crew following current recommended procedures. This situation has placed tremendous strain on the resources from the districts' maintenance group and also causes concerns related to inspection frequency and public safety.

This project investigated causes of the loose anchor bolts and proposes solutions based on site surveying, field monitoring, laboratory study, and numerical analysis. In particular, in Chapter 1 it studied how these anchor bolts were initially tightened and whether they were adequately pretensioned. Chapter 2 contained Skidmore Wilhelm testing to determine relationships between torque, rotation, and tension for different bolt diameters and grades. In Chapter 3, field monitoring was completed to quantify the torque, rotation, and tension relationships of MnDOT structures. Chapter 3 also contained testing of a laboratory specimen of a MnDOT sign structure to determine how anchors loosen during service loading. In Chapter 4, finite element modeling was completed to develop models that could be used for future parametric and fatigue studies. In Chapter 5, recommendations are made for a new specification for MnDOT structures. The objective of this project was to develop the best practical procedures using available equipment to re-tighten the loose anchor bolts so as to develop required pretension. This project will ensure that the anchor bolts will perform as designed while minimizing required inspection frequency. During testing, it became clear there was a need for cheap, digital measurements of bolt pretension. In Chapter 6, an organic sensor developed and tested for a National Cooperative Highway Research Program (NCHRP) funded project is described.

The project found that most states experience issues with loose nuts of sign and signal structures. The loose nuts are attributed to one of two reasons: inadequate tightening (under-tightened) or yielding leading to permanent deformation under service loads (over-tightened). In each case, the loose nuts can be due to an incorrect specification or contractor error. Typically, large diameter bolts are more susceptible to under-tightening, while small diameter bolts are more likely to yield and elongate under service loading. Fatigue testing of a MnDOT structure using MnDOT's previous specification for large diameter bolts resulted in loose anchor bolts due to under-tightening. The research team found that the tightening process proposed in AASHTO's specification is a sufficient alternative for MnDOT, though it requires modification in three key areas: defining snug-tight, accounting for grip length, and recommending verification procedures. Through laboratory testing and field monitoring, the research team found that there is an actual snug-tight value near 10% of yield stress. The relationship between nut rotation and bolt tension becomes linear beyond the actual snug-tight threshold. The team found that the relationships between torque, tension, and rotation beyond snug-tight for varying grip lengths can be estimated with empirical constants. Through literature review and surveying of state DOT's, the team examined verification procedures and recommends the use of a form similar to WisDOT's dt2321. The team began testing a prototype organic sensor for digital measurement.