cruise ship funnel fire

Systems Failure Investigation
Design Fault Forensics
Failure Mode Analysis

A Diversity of Cases

A third of the 6000 tubes in a large thermal recuperator failed within 24 hours of start-up. During warm-through, only a very light, temperate flow of combustion products had passed over the matrix. The Strouhal Number for the array indicated these conditions to correspond with those necessary for vortex-shedding. It was demonstrated that the frequency at which this occurred coincided with the tubes' natural resonant frequency at the prevailing temperature, causing failure under high cycle fatigue. Acting as consulting forensic engineers to the State Court of Ohio in Cleveland, we presented our failure report over the course of a five-day hearing and received commendations from attorneys on both sides.

We were appointed forensic investigators for the defence in a criminal prosecution. Its operatives had admitted to sabotaging a high pressure plant for essential oils extraction. Although they had claimed it to be dangerous, each one of them faced a long prison term - given that the owners had said it was worth a seven-figure sum. However, its price was not its value. We demonstrated that the equipment had never worked in commercial production; could not be made to work save at disproportionate expense; also, that its design and construction presented clear operational risk. The men were fined and released.

A leading manufacturer of waste heat boilers and economisers engaged us to investigate the increasing incidences of ship's funnel fires. Searches at Lloyds of London revealed little of direct interest, while inquiry into the clients' design procedures showed them to be entirely conventional. Our analysis of reporting dates reflected a history co-incident with changing practices in ship operation - particularly the burning of increasingly heavy cuts of fuel oil and much-reduced times alongside. Further calculation and investigation showed the developed profile to layer a tacky film onto tubes and flue walls at the end of one operating cycle, attracting solid smuts from incomplete combustion upon the next cool start. We demonstrated that temperatures would then rise during warm-through until the LEL was exceeded within the economiser tubestack, with ignition caused by sparks entrained in the exhaust gas stream.

Dove Thermal Engineering Limited, Uttoxeter, ST14 8QX

Telephone: 01889 569933