Turn-of-the-nut tightening of anchor bolts
Double-nut anchor bolt systems are used in the erection of traffic signal poles, high-mast luminaries, and other highway appurtenances. An absence of a tightening standard for such systems decreases the confidence in their performance under fatigue loading. Past research has shown that a tightening standard should include the development of preload in the anchor bolt sufficient to provide adequate resistance to fatigue failure. Preload should be measured by a turn-of-the-nut method.
Laboratory progressive tightening tests were performed in order to monitor the stress ranges occurring in the bolt at various locations of interest at various degrees of turn-of-the-nut tightness. Tests were performed on six diameters of anchor bolt ranging from 1 to 2-1/4 inches in diameter and two different categories of thread pitch: UNC and 8UN. Plots of stress range versus degree of tightness were developed for each test and evaluated to find the minimum degree of turn-of-the-nut at which stress range inside the nuts dropped below that outside the nuts. This shift was considered to be the principle theoretical indication of adequate performance. A fatigue test which saw failure outside the double-nut connection was set down as the practical indicator of adequate fatigue performance.
The 2 inch 8UN bolt was chosen as the critical specimen due to its overall low generation of preload during tightening tests. Theoretical testing showed that 1/24 turn-of-the-nut would guarantee sufficient fatigue performance. Two practical fatigue tests of the bolt at that tightness saw one positive and one negative failure.
After actual lab tests, finite element modeling was used to investigate the behavior of the bolt. It was found that performance did not see improvement until 1/12 turn-of-the-nut.
After all results were considered, a standard of 1/6 turn-of-the-nut or refusal of tightening by specified methods was recommended, provided a minimum of 1/12 turn-of-the-nut was achieved. This value allows for ease of measurement, sufficient tightness, degree of safety, and has been shown in past testing not to cause failure through over-tightening. However, tightening to only 1/12 turn-of-the-nut still provided adequate performance.