Afternoon all, We need to conduct repairs to metalwork on top of a flour silo. Looking at the HSe guidance, all it states is that such activities are avoided and where they can't the should be adequately controlled by a permit to work system. Has anyone had experience with such permits / RA's and aware of what controls should be specified when conducting hot work near potentially explosive atmospheres? We aren't planning to weld INSIDE the silo, rather above it near the access hatch where flour debris is visible. Obviously slightly concerned but hot work is necessary due to other safety issues.....

I agree with the above. I do have experience of permits and hotwork in an explosive atmosphere, but not flour silos, so I wouldn't consider attempting this without specialist advice. Its a recipe for disaster (there's a terrible pun there somewhere.) There must be a contractor somewhere who specialises in this kind of thing

Hi, As many people have noted, if you are not competent in this area it is worth contacting someone who is. A bit obvious maybe.... however I have experience in hot working in coal bunkers, repairs, modifications etc and the SSoW needs to be in place, again as noted. Sorry if this seems obvious or repeditive, just my 2 cents.

If the welding work is needed to the outside of the silo and all hatches and external dust deposits are removed to prevent any sparks and spatter entering the silo, is there really a significant risk? I'm also baffled by JJ Prendergast's suggestion about filling the silo with nitrogen gas AFTER removing all the flour, including dust and debris, from it. Why? surely the silo by then is an empty shell and no longer contains any combustible dust which could give rise to combustion and explosion. If flour is left inside the silo and there is a risk that sufficient heat from welding work outside can transfer by conduction to the inner side of the silo and then ignite dust deposits on the internal surfaces, it would make sense to flood the silo with nitrogen and thereby prevent the dust from combusting/exploding. On a wider note, there seems to be confusion in this case between a partially filled container of flammable liquid and one of combustible dust! If this case were about a container of flammable liquid, e.g. diesel, a small localised amount of heat from the welding work might conduct through the container to sufficiently warm up any flammable liquid residue on the inside and generate an ignitable vapour. However, as this case is about a combustible dust (flour), perhaps a notably greater amount and degree of heat is needed to reach the flour's auto-ignition temperature. My comments above are simply based on basic principles and precautions regarding control of combustion. Therefore, hopefully forum users with far better knowledge/experience of silos containing combustible fine-grained substances will see this topic and be able to respond accordingly.

Do have to agree with querying the need for introducing inert gasses into this process. In the main this would be used as part of the purging process normally associated with gas systems and would not necessarily be useful in this situation. Yes flour can cause explosive atmospheres, dependant on the concentrations that can be disturbed and become airborne and yes there was an explosion in a factory a few years ago, but Im not sure if it involved flour orsugar (could have been both.) I am pretty sure that emptying the silo followed by appropriated cleaning up of residues would sufice, if you wanted to be extra sure maybe consider static air monitoring though that can be costly. That plus the controls suggested already would be the sort of things I would expect to see in a ssow. Mind you my experiences are with coal and biomass but the theories remain the same to my mind.

I am guessing Nitrogen inerting could be a possibility, questions about the silo integrity to hold the gas without escaping (if not this could effect the working party), potential costs of inerting vs silo emptying and clean down and I suppose how long inertion could be required for? I mean heat transfer to the silo contents, could it occur? how long would it take to dissipate? All good and worthy points raised by the comments on here. I have no idea on the silo size or capacity or how easy emptying and cleaning would be.... however, and again Im used to coal and solid biomass being the hazrds, I would go down the empty and clean route, however ymmv

http://www.hse.gov.uk/food/dustexplosion.htm 2. Dust explosion hazards 2.1 A dust cloud of any combustible material will explode where: the concentration of dust in air falls within the explosion limits; and a source of ignition of the required energy is present. Containment is not always required for people to be injured and property damaged. Now why do they construct mills the way they do, and not brick walls.....

Yep this clearly requires some detailed planning. Constant nitrogen feeds would, in my experience at least, be an expensive way of controlling this, as well as potentially providing a positive pressure into the silo (again a generalised statement as I have no idea of silo design) potentially causing a dust cloud. Meh this is all general thinking on my part as I say I have no idea on the designs involved, but you might want to do some research on confined and unconfined dust cloud explosions, I could probably dig some of my fire risk management stuff out if you wish.

Agreed JJ :) For my 2 cents I would go with the empty clean down, removing the hazrd entirely, ensure the working area is safe, as I said general ideas and the afairp principle will be the guiding factor

bob How will a fire watch be effective? If there was an explosion, it would happen too quickly for the fire watcher to be effective. Don't think it would be a small fire, taking some minutes to get bigger. Plus the fire watcher might get killed/injured in the explosion. As per my previous post, fully accept the silo is not likely to be gas tight. The intention would be to greatly reduce the level of oxygen in the silo to render an atmosphere incapable of supporting combustion. Neither would it be the intention to pressurise the silo, as it probably isn't pressure rated. The most practical solution is to probably use the water technique. I do sometimes find it amusing on this forum, when posters ask for advice/opinion on how a problem/mitigate risk. The usual responses are - 'You need to do a risk assessment' Or 'See reference xxx' etc Or quote extracts of regulations Often the original poster already knows he has 'to do a risk assessment' and knows about Google and the HSE website. But whats the practical solution? Accept the information is often not always detailed enough to give a detailed answer.

JJ, I think the purpose of the fire watcher would be to ensure that a- the heat transferred through any plating doesnt cause hazards the other sides and b- to ensure checks are made after work is complete. I believe the poster making this point mentioned vaccuuming to remove flour dust, effectivly removing the combustible materials or potential explosive atmospheres. Apologies if this comes across as egg sucky. With regards to lowering the o2 levels to a level where combustion cannot start using nitrogen, again a distinct possible control measure, but I would suggest that the cost of continuously maintaining the nitrogen levels, plus continual atmospheric measuring, added to the potential extra costs that might be necessary to protect any working party should it be necessary could well exceed the costs of simply emptying and cleaning out the silo, then providing a one off atmospheric report proving that the entire work area is safe with regards fire and or explosion, at least as far as is reasionably practicable is concerned.

Dimond Agree, the nitrogen method is probably more expensive/less practical than cleaning and washing out. I do believe all options should be raised for discussion and screening, however - given the potential serious outcome of getting this sort of job wrong. As ever, there is not enough information in the original post to come to any hard decisions

I have worked with flour silos many times. I suppose this is located outside? I would ensure the silo is emptied and the bottom and top hatches are opened for at least one hour and a dust monitor is used before the welding work starts. I don't think it is a particularly hazardous task in terms of explosion or fire (providing the silo is emptied and properly aired). I would be more concern with regards to the working at height issues here. There won't be much up there to catch fire following the welding task, nonetheless, I would give it time for surfaces to cool down and the silo to be cleaned inside before filling up and putting back in action. If you have a contractor doing this, ask them to provide you with a RA and method statement and do your own too, but like I said, pay attention at WAH.