The Hierarchy of Control in Practice

02 Feb 2024 at 19:02 // under Work Safety

This post is based on an event that occurred several years ago now, after I had become a control engineer but when I had been seconded back to an operations role during a shutdown.

Knowing the hierarchy of control is one thing but it is useless if we forget to apply it. I fear that it is all too easy to just reach for the control measures that we are used to using rather than taking a fundamental look at the task.

In the middle of a nightshift during a shutdown, I was approached by one of the supervisors and asked to prepare a risk assessed operating instruction to cover a leak test of a heat exchanger. Leak testing is regularly carried out as part of returning a piece of equipment to service. It is not a full pressure test, which is usually carried out using a liquid such as water or lube oil. A leak test is to ensure that the joints, such as flanges, have been connected properly and will not leak when operating. Another task that needs to be carried out before reintroducing hydrocarbons to a piece of equipment is air freeing. This is where any air that is in the equipment is purged and displaced by an inert material, usually nitrogen, so that no flammable or explosive atmospheres are generated within the equipment during commissioning. These two steps are often combined. The pipework or vessel is pressured up to about 3 bar¹ and held while all the joints are checked. Checking involves squirting some ‘leak detection fluid’ onto the joints and looking for bubbles that are generated by any leaks. If bubbles are spotted, the joint will need to be fixed and retested before commissioning can continue.

What made this leak test different from normal was that the joint to be checked was inside the shell of the heat exchanger. To check for any leak, someone would have to put their upper body inside the exchanger. This converts the job into a confined space. And as a result an extra level of risk assessment and method statement is required.

To be fair, the level of risk is still pretty small. Even if the joint was leaking, the volume of gas escaping would still be very small. The enclosed space is also pretty open, allowing plenty of air circulation and as long as someone else is standing beside them, they could easily pull their legs and get the person back into the atmosphere without any risk to themselves.

As I sat down to write the procedure, I instinctively went with the methods and control measures that I know including the use of a gas monitor and standby attendant. But then it struck me that I wasn't actually applying the hierarchy of control. The best thing to do is to eliminate the hazard. And in this case I could. To do a leak test, you don't need to use nitrogen. It is normally more convenient to use nitrogen because you have to use it for air freeing later, and it doesn't normally increase the hazard. But in this case, there is an increased hazard and an alternative method of testing is available in the form of using compressed air.

If compressed air is used, it doesn't matter if the joint leaks, it is just air. The hazard has been completely eliminated.

It is not often we get to completely eliminate the hazard. As a result, I think it is easy to forget to consider the possibility. It is easy to simply use the tools that we get used to using, even if they are further down the control hierarchy.

1. Assuming the vessel is rated to 3 bar or higher ↩

Go Top