PSI - Issue 45

Aditya Khanna et al. / Procedia Structural Integrity 45 (2023) 12–19 Khanna and Young / Structural Integrity Procedia 00 (2019) 000 – 000

13

2

1. Introduction Vibration-induced fatigue of process pipework and the resulting loss of fluid containment poses a significant structural integrity risk for asset owners and operators across many industries. Aside from production downtime and the cost of remediation works, there are safety and environmental concerns as well (Matta and Szasz, 2018). Piping vibration modes are classified as either bending modes (typically up to 300 Hz) or shell modes (typically above 300 Hz). Smallbore fittings, such as pressure vessel nozzles, are generally stiffer than mainline piping and their bending modes can occur at natural frequencies well above 300 Hz. The present study, and the vibration criteria reviewed herein, pertain to the bending (transverse) modes of vibration. The objective of the present work is to identify vibration criteria that are most suitable for the field-based vibration assessment of pressure vessel nozzles. This sub-category of small-bore connections requires special attention due to the flexibility of the parent pressure vessel (i.e., the nozzle base) and the relatively stiff geometry of nozzles compared to general pipework. Fig. 1 shows an example of a pressure vessel nozzle and its fundamental (first) mode of vibration.

Fig. 1. First-mode transverse vibration of a pressure vessel nozzle. The inset sketch shows the typical weld detail at the nozzle base (combination fillet and bevel weld).

1.1. Vibration assessment guidelines Over the decades, several empirical and semi-empirical guidelines for acceptable piping and small-bore fitting vibration have been proposed for the field-based assessment of vibration severity. Vibration analysts rely on these industry- accepted guidelines for “screening” vibration datasets, i.e., to identify areas of high vibration (high dynamic stress) on in-service process pipework. Only a brief recap of the industry guidelines relevant to the present work is presented here. Background information on the theoretical basis and underlying assumptions of these industry guidelines is provided in many other works, e.g., Jacimov ic and D’Agaro (2020) and Bifano et al. (2018). One of the earliest acceptable pipe vibration limits was by the Southwest Research Institute (SWRI) in the 1970s based on statistical data on piping vibration and failures (von Nimitz, 1974; Wachel et al. 19 90). The “Design” line for allowable vibration amplitude, in mils P-P, was set to (250 ⁄ ) 0.5 , where is the dominant vibration frequency in Hz. The “Marginal”, “Correction”, and “Danger” lines were defined as approximately 2x, 4x, and 10x the “Design” line , each corresponding to a lower margin of safety than the previous (Fig. 2a). Pipe vibration levels exceeding the “Danger” line strongly correlate with fatigue failures (Wachel et al., 1990). While the vibration limits in the empirical SWRI design chart are expressed in terms of displacement amplitude, vibration velocity is the preferred measurement quantity for the initial screening of field measurements. This is because the dependence of the dynamic stress on the vibration velocity at resonance is the least sensitive (compared to displacement or acceleration) to the pipe geometry and boundary conditions (Norton and Karczub, 2003). For this