# 1995 Edinburgh
The failure of a pipe by rapid crack propagation (RCP) can, by its nature, be highly desmctive and must be avoided. Research over the past ten years has concentrated on gas pressurised pipelines and has resulted in a greater awareness and understanding of the problem. Recent attention has focused on RCP in water pipelines, raising two new questions : how does the change in pressurising fluid affect RCP, and how well do plastic pipe materials, other than polyethylene, resist RCP failure? The small-scale steady-state (S4) method has been used to test thoroughly both medium density polyethylene (PE-80) and unplasticised PVC (PVC-U) water distribution mains pipe. Less detailed testing was also canied out on some pipe materials with a 50 mm external diameter. These again included PVC-U, and some materials with a greater resistance to chemical attack, namely poly (vinylidene fluoride) (PVDF), polypropylene homopolymer (PP-h) and copolymer (PP-c). This work studies the effect of varying the air volume in watedair pressurised pipe in which small air volumes simulate service conditions in water pipes. The results were similar for all materials, showing a rapid decrease in critical pressure with increasing air volume up to 20% air. The critical pressure then remained unchanged as the air volume increased up to the fully air pressurised case. Dimensional analysis is proposed as a method of understanding pipe fracture. The results suggest that, with a small amount of entrapped air, pipe fracture is governed by material modulus and diametedthickness ratio and is independent of fracture resistance.