Mike Johnson and Steve Critchlow answer your #BurningQuestions
Every month, we invite experts from across the gas industry to answer your most pressing questions. This month, we’re joined by two of the most respected and knowledgeable figures in gas safety, Mike Johnson, VP & Senior Principal Consultant at DNV, and Steve Critchlow, Principal Gas Engineer at the Health and Safety Executive Science Division. With decades of experience between them, Mike and Steve share their insights into carbon monoxide safety, incident investigation, and the evolving risks and opportunities as we move towards a hydrogen future.
Mike Johnson, VP & Senior Principal Consultant
Mike has over 46 years’ experience in hazards research and incident investigation, originally joining British Gas Research & Development (now part of DNV) in 1978. In 1984 he took over responsibility of carrying out large scale experimental research of vapour cloud explosions. This work included managing several large-scale joint industry projects studying explosion hazards and mitigation techniques.
He has investigated over 60 explosion incidents mostly related to domestic explosions in the UK but including major vapour cloud explosions at Buncefield, UK in 2005 and Jaipur, India in 2009.
He is now VP Senior Principal Consultant within DNV Spadeadam Research and Testing, providing technical governance for a range of projects. This includes studies of the hazards associated with hydrogen. He has also been a technical lead in a DNV led consortium providing technical advice to the Department of Energy Security and Net Zero (DESNZ) Hydrogen for Heat programme.
Steve Critchlow, Principal Gas Engineer at the Health and Safety Executive Science Division
Q: Carbon monoxide cannot be detected by human senses, and visual signs are not always reliable. Without testing the air or appliances, how can First Call Operators confirm whether customers are exposed to CO?
Steve: The role of the First Call Engineer is to make the situation safe. This is an important task which must be performed diligently. The emergency service provision by gas networks gets a qualified operative there quickly to make safe by isolating the fuel source, to issue warning notices to the user, to advise seeking medical help, and where appropriate, to notify HSE there’s been an incident. This should then stimulate an investigation by a suitably qualified gas engineer.
Currently the only reliable way to confirm CO exposure is testing by a qualified medic, and this is one reason why prompt attendance at a hospital is important.
It is worth keeping in mind that CO also comes from solid fuel such as wood, coal and biomass, oil, and petrol driven engines such as in generators. In instances where this is the source it's likely the gas network will not be involved. Therefore messaging is required to ensure those exposed via these sources also seek medical attention and get their installations checked by a competent person.
Q: With so few RGEs holding the CMDDA1 qualification, and difficulties finding them on the Gas Safe Register, how can customers access support after a CO alarm sounds? What options are available for tenants, who may be reluctant to approach their landlord?
Steve: When a CO alarm sounds it is permitted that a gas engineer who is qualified to work on all the appliances present can attend to carry out all the routine safety checks that they are trained to do. If they do not find a reason for the CO alarm sounding, which can include things such as the alarm failing or the gas user being mistaken about the source of the “beep” in addition to gas safety defects, then a CMDDA1 investigation to check the air quality, and in particular to look for CO entering from elsewhere, may be required. This approach speeds up the process of getting a qualified Gas Safe Engineer into the property. This is important as we know people will be tempted to continue to use appliances if there’s a significant delay in attendance.
One area of concern is the increase in multi-fuel homes. If there is a gas boiler and a solid fuel stove present, for example, it is important the attending specialist does not declare the situation safe after only examining the fuel for which they are qualified. An investigation of causation must include qualified inspection of the alternative fuel appliance and its flue.
In a rented property the legal duty is on the landlord to maintain and annually inspect the installation. If for any reason the tenant does not wish to approach the landlord they are able to contact any suitably qualified Gas Safe Engineer directly, but they must be aware that in this situation the tenant may well need to cover the cost of the services.
Q: What common factors do you see repeatedly in gas-related incidents and accidents?
Mike: In terms of more serious explosions in domestic properties, a little more than a third result from leaks outside the property, generally from older metallic mains or services. Inside properties, appliances account for about half of the leaks with the rest being from pipework and meter. A relatively small number of incidents are due to deliberate interference.
Steve: I would add that it's quite common, sadly, for people to tell us they’ve been smelling gas for some time but didn’t report it. As an industry we perhaps have more to do to make people aware they can report to the gas emergency service provider without cost and without judgement. No one will be critical if it turns out not to be gas.
Q: How do you think the industry can improve the reporting and learning process from near-miss events?
Steve: The main system we use for collecting incident data is RIDDOR. For gas 11(2) requires the reporting of immediately dangerous installations due to workmanship which haven’t yet caused an incident. My perception is that there is very wide non-compliance with this by registered gas engineers. This is a shame and a problem because without supporting data it's difficult to make the changes required for improvement.
Q: In your experience, what role does human error play compared to technical failure in gas safety incidents?
Mike: In terms of a direct cause, very little.
Steve: A gas release or a release of CO is always facilitated by a technical engineering failure, such as the breaking of a buried cast iron pipe, or the separation of a flue.
In terms of CO, modern gas appliances have many safety features, and if they are installed completely and compliantly they should stay safe throughout their serviceable life subject to competent inspection. However, human factors such as the decision to employ an unqualified installer or the decision to DIY repair can make the installation unsafe. In addition we do sadly find engineers who are fully qualified but who do an incompetent job. This must be because they have taken a decision, either consciously or subconsciously to cut corners despite their training.
With gas related fires and explosions, human factors would include things like a decision not to report a suspected smell of gas, or a failure by an engineer not to solder a pipe or carry out a tightness test when required to.
In my opinion it is clear that if we could find a way to remove human failings the number of gas related incidents would be dramatically reduced.
Q: How do you see the introduction of hydrogen into the network affecting incident risks and investigation methods?
Mike: There has been a substantial amount of work carried out to show that with additional safety measures, the risk to people from hydrogen in the gas network would be similar to the current risk with natural gas, which is already extremely low. Investigation methods would remain largely the same, though some testing may be required in order to allow interpretation of forensic evidence.
Steve: Hydrogen will introduce some new risks, but will reduce others; the obvious example being carbon monoxide. As Mike says, industry and the regulator have done huge amounts of work to ensure the gas industry will remain safe if hydrogen is introduced.
Q: What advice would you give frontline engineers to reduce the likelihood of domestic explosions?
Mike: Follow well defined procedures for emergency response and err on the side of caution.
Steve: Be aware that the risks you learned about on day one still exist and are waiting to catch you out. Do not let familiarity make you less diligent. Treat each job as though you are being examined, and carry out all the checks required by your procedures, standards and regulation.
Q: Looking back at large-scale incidents like Buncefield or Jaipur, what lessons are still most relevant for the gas industry today?
Mike: Understand the condition of those measures that prevent major accidents. This can be equipment, people and procedures. Understanding that if some things are failing, then other risk control measures need to be applied is important. Very simple and zero cost measures would have prevented the Buncefield incident. It is the reality of the condition and performance of these barriers to a major accident that matters, not what we expect them to be doing.
Q: From the cases you’ve investigated, what are the most frequent causes of domestic gas incidents?
Steve: In terms of explosion I would say the three most common causes, in no particular order, are; failed cast iron pipes outside the house, the use of butane in the manufacture of illegal drug related products, and the failure of single stage LPG regulators. The causes are many and varied though.
We have had significant spikes of incidents caused by specific issues like the Beko/Flavel gas cookers with sealed grill enclosures. However, generally with CO, the main causes in recent years are the separation of a flue (sometimes through lack of support, sometimes through damage by a third party), and the incompetent alteration of boiler air/gas ratio valves. Blocked chimneys on open flue gas fires remains a significant cause.
Traditionally with Open Flued appliances a lack of regular maintenance was the major cause of CO issues.