||In recent years, there have been great advances in the modeling of unsteady friction in transient flows. It is likely that in the near future, unsteady friction models will be applied to practical, real-life problems. However, the practical use of unsteady friction models requires justification. Issues which require exploration include: i) under what conditions are unsteady friction effects large enough to warrant the use of an unsteady friction model; and ii) for what applications is the use of unsteady friction highly important? This thesis investigates the relative importance of unsteady friction in pipe systems with external flows. External flows represent any inflow/outflow elements in a pipe system, such as demands, leaks, or branching subsystems. Energy-based investigations, performed with a quasi-two-dimensional, unsteady friction, transient flow model, show that in general, unsteady friction is highly important for small external flows, but negligible for large external flows. The result for small external flows shows that for pipes with small leaks, unsteady frictional effects may be larger than leak-induced effects. This shows that transient-based leakage detection techniques should consider unsteady friction. The damping rate leak detection method, which was originally derived assuming quasi-steady friction, is investigated with unsteady friction effects included. The investigation shows that the method still works when unsteady friction is important, as long as an unsteady friction model is used to determine the frictional dissipation rate of the system. The damping rate method was originally derived under highly restrictive assumptions, which are difficult to satisfy in practice. These assumptions are systematically relaxed, and the method is found to remain valid as long as the initial flow perturbation is rapid and impulsive, and the pipe system is a simple reservoir-pipe-valve setup.