Scale of the problem

Boilers operate under the premise that water boils at higher temperatures when under pressure. Should the integrity of a boiler be compromised by exceeding the design limits, or because of mechanical failure, the results can be catastrophic.

The water in a boiler is in a liquid state only because it is under pressure. As soon as the pressure is removed the water will flash into steam in adiabatic reaction. This steam, while at a lower pressure, will expand at an exponential rate, thus the fragments of an exploding boiler are propelled at horrifying speed.

When failure occurs...
In March 2001 a calorifier used to provide domestic hot water ruptured during operation in a factory.

The system had been out of service for an extended period and scheduled for demolition for five years. As it was due to be scrapped it had not been descaled, but a need for extra hot water saw it returned to service. Shortly afterwards it ruptured under pressure. The vessel tore along a riveted longitudinal seam and around both end caps. The force of the explosion moved the vessel about two metres off its foundations and severely damaged adjacent equipment and parts of the building in which it was located.

The inspector's report noted that the bends at the ends of its tube bundle had what appeared to be long-term minor cracking and evidence of more recent sudden cracking. It was surmised that this recent cracking caused an overpressure that could not be handled by the relief valve, especially in what appeared to be its impaired condition resulting from a long-term build-up of scale around the safety valve seat and disc, and the vessel ruptured. A later report concluded that the disc and valve seat could not be separated because of scale build-up and the valve did not lift under a test pressure of 12 bar.

A further illustration of the dangers of scale build-up is well documented in an explosion of a calorifier in a modern single-storey office building. The calorifier rocketed through the reinforced concrete floor of the building and through the roof. The blast was of sufficient velocity to cause considerable damage to the building, computer equipment, basement plant and a telephone exchange.

The calorifier had ripped apart at the bottom at, or near, the shell-to-endplate seam. It was estimated that an explosion force equivalent to around 10·4 kg of TNT had been created to cause the damage. It was established that the calorifier had been out of use for a period and the flow and return valves on the shell had been closed. On the day of the accident steam had been leaking into the coils. Over a period of 18 hours the temperature, and consequently the pressure, of the 2727 litres of water in the calorifier gradually increased until the vessel exploded at what was later calculated to be a pressure of approximately 4·5 bar, compared to its normal operating pressure of 1·6 bar.

A vent pipe was fitted to the system, but was isolated when all valves on the calorifier shell were closed. This only left the safety valve to relieve the pressure, which failed as scale had accumulated around the valve head and its seat.

Following the accident, laboratory tests were carried out to simulate conditions where a safety valve can be prevented from lifting at its set pressure due to adhesion between the valve head and its seat. Experiments were conducted on various valves at different temperatures. Water obtained from the site mains was dripped into safety valves and allowed to evaporate. Scale was thus formed and on one occasion, a valve, which had been set and tested at 2·4 bar, was scaled to its seat after only 40 days. Under subsequent hydraulic test the safety valve operated with a loud report at a pressure of 6 bar. The inspectors concluded that it was likely that the safety valve had wept. In the warm environment of the plantroom this would have produced the scale, which had built up in just a few months.

Following their findings the inspectors conducted trials on other safety valves taken from heating boilers and systems in different parts of the country. Of the total number of valves checked approximately 2% showed significant sticking.

Regular testing of safety valves is essential and it is imperative that ongoing scale prevention is carried out on all boilers and calorifiers to avoid accidents.