The Weekly Reflektion Week 10 / 2020

Do you know how your utility and process systems might be connected? How easy would it be to get hydrocarbons into a utility system and perhaps into a non-classified area? Do your HAZOPs specifically consider this eventuality?

An incident on a facility in the Norwegian sector highlights the importance of separation of process and utility systems. Poor design or incorrect operation can lead to hazardous materials in systems that are not designed for hazardous materials. When this happens, there is a significant risk of a Major Accident.

During the erection of some scaffolding on the facility the scaffolders inadvertently closed the isolation valve on the air supply to an actuator on an emergency shutdown valve (ESDV). The emergency shutdown was a blowdown valve and designed as fail open. That is, the valve moved to the open position on loss of the supply to the actuator. The opening of the ESDV led to a chain of events that eventually led to a release of 50 litres of oil and 200 kg of gas into the atmospheric vent system and then through the atmospheric vent located on the flare tower.

The incident could have been prevented if the isolation valve had been locked open or the handle removed. This is one of the recommendations from the incident investigation.

The incident highlights the hazards associated with connection of live process systems and utility systems. The atmospheric vent system is normally only used during maintenance and inspection activities on major vessels and should be physically isolated from process systems when the process systems are in operation. The atmospheric vent system is an extensive system that connects process equipment in different areas together. Any incorrect design or incorrect operation could lead to hazardous materials in any of these areas. No single failure should lead to hazardous materials ending up in a non-hazardous area.

The incident reminded me of an incident on an installation in the UK in the 1980s where gas was released in the accommodation module through the toilets. The toilets used sea water for flushing, and the sea water was supplied from a seawater main that encompassed the whole platform. Sea water from this main was also used for cooling high pressure gas in a shell and tube heat exchanger. The heat exchanger was protected from overpressure on the shell side (seawater side) by a bursting disc that released the gas to the flare. One of the tubes ruptured on the heat exchanger and the bursting disc ruptured. However, gas at high pressure started to flow back into the seawater main through the seawater input to the exchanger. There was no non-return valve that prevented the backflow of gas into the seawater system. Gas filled the seawater system and eventually found its way to the toilets in the accommodation module. How surprised were the platform personnel when gas started to flow around their vital parts? How lucky were the platform personnel when the gas did not ignite, and no permanent damage was done? How lucky to be sitting where they were when the %#¤& hit the fan.

We are aware of incidents where hydrocarbons have found their way into the instrument air system, into the plant air system, into the seawater system, into the nitrogen/inert gas system and into the atmospheric vent systems. These utility systems are extensive and could lead to hydrocarbons being released in other areas including non-classified areas where the ignition control requirements are less stringent.

Do you have a good understanding of how your process systems are isolated from your utility systems and how the isolations may be compromised? Are you aware of the operations that are carried out where utility systems are connected to process systems?

The most useful tool for this purpose is the HAZOP. Next time a HAZOP is planned then include an assessment of system segregation and assurance that physical isolations are in place. And if you are concerned, organise a new HAZOP.