Confined spaces

This talk is about a space that is enclosed, or largely enclosed, where someone could be seriously hurt by what is in the air or by what can flood in. HSE's definition is specific. A confined space is both enclosed or largely enclosed and has a reasonably foreseeable specified risk to workers of fire, explosion, loss of consciousness, asphyxiation, or drowning. It can be small and restrictive. It can also be large. HSE's example of a large one is a grain storage silo. A loft, a cupboard or a plant room is not a confined space just because it is awkward. It becomes this talk when it is enclosed and one of those risks is reasonably foreseeable. I read HSE's introduction, the legislation page, and leaflet INDG258 on 20 October 2026.

The danger is not the squeeze. HSE says working in a confined space is dangerous because of noxious fumes, reduced oxygen, or a risk of fire. Other dangers include flooding, drowning, or asphyxiation from dust, grain or another contaminant. INDG258 describes the same idea as any space of an enclosed nature where there is a risk of death or serious injury from hazardous substances or dangerous conditions, for example a lack of oxygen. Some spaces are obvious. INDG258 lists storage tanks, silos, reaction vessels, enclosed drains and sewers. Others are easier to miss and can be just as dangerous: open-topped chambers, vats, ductwork, and unventilated or poorly ventilated rooms. HSE says it is not possible to list every confined space. Some places become confined spaces when work is carried out in them, or during construction, fabrication, or later modification. A chamber that was safe to look into yesterday is not a method for today.

The law splits into two steps I want read out. The Management of Health and Safety at Work Regulations 1999 require a suitable and sufficient assessment of the risks, so you can decide the measures. For this work, INDG258 says that assessment usually takes in the task, the working environment, the materials and tools, whether the people doing it are suitable, and the arrangements for emergency rescue. You may need a competent person to help, and the people doing the work have to be trained and instructed. If that assessment finds a risk of serious injury from work in a confined space, the Confined Spaces Regulations 1997 apply. HSE's legislation page sets out the three duties: avoid entry, for example by doing the work from outside; if entry cannot be avoided, follow a safe system of work; and put adequate emergency arrangements in place before the work starts.

Avoiding entry is the duty, not a preference. Ask whether the job has to be done in there at all. INDG258 says you can modify the space so entry is unnecessary, or do the work from outside. Its examples are clearing a blockage from outside, and using a remote camera to inspect a vessel. On our work, that question comes before anyone drops into a tank, a sealed void, a duct, a sewer, a manhole or an open-topped chamber. If we can see what we need to see, or clear what we need to clear, from outside, we stay outside.

If entry cannot be avoided, the controls are for this space, not for a generic tank. HSE's introduction says: if there are harmful fumes, consider how to ventilate them or remove them; if liquids or gases could flood in, establish whether the valves can be locked shut; if there is not enough oxygen to breathe properly, provide breathing apparatus or ventilate the space to increase the oxygen before anyone enters. INDG258 says mechanical and electrical equipment that could otherwise operate has to be isolated, and pipework has to be isolated if gas, fume or vapour could enter. In every case, check that the isolation has worked. The person who goes in has to be capable, and trained in the work and in the emergency equipment. Do not do this work unless it is essential. Do not ignore the risk because the space was safe on another day. Do not let the next person in until you are sure it is safe. Everyone involved needs to be trained and instructed in the safe system, and that system has to be put into practice, not left on a sheet in the van.

HSE's introduction gives an example I am not softening. Two men went into a closed crane compartment to look at a hydraulic fault. Within a minute, one had passed out and the other felt lightheaded but got out. Two others went in to help and were overcome. The emergency service got the three men out. Two of them died. Water in the compartment had caused rusting, which depleted the oxygen. HSE says that if the oxygen had been checked, the space could have been ventilated and the deaths avoided. The point for us is the rescue, not the crane. You need to know the person inside is all right and has not been overcome by fumes. You need a way to get them out. HSE says it is not enough to rely on the emergency services for that.

What we decide before anyone goes in today: is this space enclosed, and is fire, explosion, loss of consciousness, asphyxiation or drowning reasonably foreseeable? Can the job be done from outside? If not, what is isolated, what has been done to the air, who is trained to go in, who is staying outside, and how we get a person out without waiting only on the emergency services. If we cannot answer, we do not go in.