Hacker News Reader

Comment by jacquesm | Hacker News Reader

Welding, roof work, grinding. Those three are responsible for a good chunk of all fires. I've done quite a bit of all three and have to confess that once or twice I was lucky rather than smart to have no bad effects from a very small mistake. When grinding, all it takes is a rag used to degrease something days before at 30' to set it on fire. When welding you really want to keep a very good idea of what is on the other side of your weld at all times. Surprise: a box member of a car filled with PU foam. I really never saw that one coming. And finally, when working on a roof a friend of mine did not properly calculate in the effect of an exothermic reaction in a vat of resin exposed to the sun. Close call that one, averted by denying oxygen to the already burning vat.

This one is on a completely different level though, and I'm sure that the welders did not live to tell the tale. Even so, before you go and claim they were stupid you have to take into account that this is Beirut, not exactly a place where the local OHSA is going to beat down the doors to ensure everything is done safely and by the book, that in a harbor there are always lots of dangerous things in close proximity and that they may have taken all possible precautions and still ended up drawing an unlucky card.

It really doesn't take much.

View on HN

It really doesn't take much.

It really doesn't. You don't even need a grinder. Oily rags can ignite spontaneously. People usually laugh at the phrase "spontaneous combustion" because it's so often associated with "spontaneous human combustion." But actual "spontaneous combustion" does happen in certain (not uncommon) circumstances. See: https://www.nfpa.org/-/media/Files/Public-Education/Resource...

And what makes things even scarier, is realizing that buildings, ships, etc. that are either under construction or undergoing maintenance are more "at risk" because alarm systems, automatic fire suppression systems (eg, sprinklers), etc. are often times not in place yet, or disabled, while work is going on.

This is one reason it's so common to see a building complex that is under construction burn to the dirt if it catches fire, as opposed to a finished building where you might get a "room and contents" fire. No sheetrock, just miles and miles of exposed wood, no sprinklers, no alarm - recipe for disaster.

> Oily rags can ignite spontaneously.

That happened to a friend of mine. He was doing some woodworking and threw a linseed-oil soaked rag in the trash, then went to work. He returned home to his apartment on fire and a bunch of firetrucks blocking the driveway. Linseed oil apparently spontaneously combusts at 120 degrees and generates heat as it dries, so it can cause a fire in normal 70-80 degree weather.

Woah. I feel like I might have heard about Linseed oil’s propensity to spontaneously combust but forgotten about it. Thanks for the reminder.

Woodworkers are warned all the time about this. Lots of fires caused by hobbyists finishing furniture that aren't aware of it.

> buildings, ships, etc. that are either under construction or undergoing maintenance are more "at risk" because alarm systems, automatic fire suppression systems (eg, sprinklers), etc. are often times not in place yet, or disabled, while work is going on

Reminds me of the loss of the Normandie / Lafayette: "At 14:30 on 9 February 1942, sparks from a welding torch used by Clement Derrick ignited a stack of life vests filled with flammable kapok that had been stored in the first-class lounge. The flammable varnished woodwork had not yet been removed, and the fire spread rapidly. The ship had a very efficient fire protection system, but it had been disconnected during the conversion and its internal pumping system was deactivated." https://en.wikipedia.org/wiki/SS_Normandie#Fire_and_Capsizin...

This kind of thing is surprisingly common. The recent fire on the USS Bonhomme Richard had a similar issue.

Since the ship was in maintenance, on-board fire-suppression systems had been disabled, delaying the onset of firefighting efforts, according to Admiral Sobeck

https://en.wikipedia.org/wiki/USS_Bonhomme_Richard_(LHD-6)#J...

Another one that didn't get as much national attention was the fire on the SSG Edward A. Carter, Jr at the Sunny Point Military Ocean Terminal back in 2001.

According to a subsequent Coast Guard report, the ship’s second assistant engineer started a transfer of about 20 tons of heavy fuel oil from the port and starboard overflow tanks to a central settling tank. The transfer was left unsupervised other than by automatic equipment.

“Their electronic system measured the tank levels and sounded an alarm if the preset levels were exceeded,” Sledge said. “If you are starting to overfill the tanks, it sounds a warning tone.”

Unfortunately, because cables to several tanks had become contaminated with fuel oil, false alarms had become a repeated nuisance. The easiest solution was to simply turn off the alarms.

https://web.archive.org/web/20120723160038/http://www.firewo...

> easiest solution was to simply turn off the alarms

This is so common... it's like the holy trinity of "Seconds from Disaster"

The one time you need them...

Or the very recent fire on the USS Bonhomme Richard.

Speaking of spontaneous combustion - ammonium nitrate will spontaneously ignite in the presence of some impurities.

Ammonium nitrate plus zinc powder is a mixture that only needs a drop of water or just sufficiently humid air to ignite.

Another one is just plain ole organic matter, like mulch. Back in my days as a firefighter, our district included the county landfill. After one of the hurricanes that came through, they had gone around collecting all the downed trees, and chipped them up into mulch, and located this HUGE pile of mulch at the landfill, where county residents could come and get free mulch / wood chips for their gardens or whatever. Sounds great, right?

Except it turns out that a huge mountain of decaying organic matter, left exposed to the southeast NC sun in July and August will periodically catch fire spontaneously. We went out there a few times to put that mess out. Not an easy task. You have to bring in heavy machinery to dig into the piles in order to get to the seat of the fire...and of course the landfill was nowhere near a hydrant, so we had to call out half the tankers in the county to set up a tanker shuttle... uuugggh.

The sun is actually not much of a factor in there. What makes this happen is that the mulch rots which is an exothermic process and it insulates very good as well. So the core temperature of pile will slowly creep up until you reach the ignition point. The same will happen to bales of hay that are taken from the field too wet.

What makes this happen is that the mulch rots which is an exothermic process and it insulates very good as well. So the core temperature of pile will slowly creep up until you reach the ignition point.

Absolutely true. But while I wouldn't even be able to begin quantifying the degree to which the radiant heat from the sun is a factor, I strongly suspect that it is a factor, if for no other reason that the observation that we only went to those fires in summer months. But, to be fair, there could be a million other confounding factors.

Whether or not a given system ignites on a given day depends on many things: ambient temperature, wind, humidity, presence of accelerants, yadda, yadda, yadda.

I suspect this is one reason why it's so hard to get people to take fire safety seriously: that pile of oily rags in the corner? It's not going to catch fire. This time. Or next time. Or next time. And so on... until it does. So people get complacent because "hey, the oily rags never caught fire before, why would I expect them to today?" Disconnect the sprinkler system? Sure, why not - the building isn't going to catch fire. This time...

I know the former head fire marshall of a major European city (2M inhabitants). The stories about root causes have made me extra fire safety sensitive and it is a continuous source of frustration to me that others don't take this stuff as serious. Cheap power adapters are somewhere near the top of my paranoia list, anything rechargeable battery operated a close second.

Fun fact, the metabolism of the sun produces about as much energy per unit volume as a pile of mulch.

bales of rotting wet hay is where barn fires come from.

Yep. They also sometimes spontaneously combust in the field.

Yes they do. We went to more than a few hay bale fires as well. Those are also a pain in the ass, and for similar reasons.

So what's the procedure like? Pull them apart and then put it out? I'd assume that if you just wet them from the outside you are just setting yourself up for a speedy return to the same location.

So what's the procedure like? Pull them apart and then put it out?

Exactly. Besides trapping heat in, the hay also forms a fairly effective barrier against water getting inside. If you don't rip the bale apart to fully extinguish the seat of the fire, you just wind up right back out there.

You can see the process underway in this video, where a bunch of hay bales caught file while on the back of a truck being transported. Scroll a few minutes into the video and you can see where they brought a front-end loader out there to start ripping the bales apart.

Using a "wetting agent"[1] also helps, and some departments will use store bought dish soap, like Dawn, for that purpose (purpose made firefighting foam is kinda expensive).

https://youtu.be/z1_pAH34wAo

[1]: https://en.wikipedia.org/wiki/Surfactant

Interesting video, especially with a drone over so much hot air. Amazing how much land is already gone. And the truck itself is just a memory while plenty of the bales of hay are still sitting around unharmed, you can really see the 'R' value at work there.

It almost looks as though the fire started in the truck itself or on one of the rear brakes, was there a report on the cause of the fire in this particular case?

edit: amazing coordination as well, at one point working on the fire in four different places.

It almost looks as though the fire started in the truck itself or on one of the rear brakes, was there a report on the cause of the fire in this particular case?

I'm afraid I don't know. :-(

Spreading woodchips dumped shortly before first snow some years back, I noted on a well-below-freezing day that the centre of my woodchips pile was straming & smoking.

Chips melted through a few inches of snow when laid down on garden paths, and remained clear of snow for at least several days.

No explosions.

You have to bring in heavy machinery to dig into the piles in order to get to the seat of the fire

Wouldn't that expose more of it to oxygen and thus increase the fire's strength? Or does it generate oxygen itself?

Wouldn't that expose more of it to oxygen and thus increase the fire's strength?

Yes, but there really isn't much else in the way of practical choices. If you had some 100% bullet-proof way to seal the fire off from any oxygen, then eventually it would go out and cool below it's ignition temperature. But "bury it in sand (or concrete)" isn't a practical solution for the average fire. Beyond that, you could play around with piercing nozzles and more wetting agents / foam and what-not and try to avoid digging the pile up, but generally speaking those things would be too time consuming for the typical fire department.

Consider that any time you're on a call, that apparatus and crew are not available to respond to other, perhaps more urgent, calls. So fire departments typically try to avoid spending excess time on the scene of a relatively low hazard fire like a mulch pile. You can't just ignore it completely, because embers from that fire could then spread and cause another fire, etc. So you have to show up and deal with it, but you want to get done as soon as possible.

A private / industrial fire brigade might take a different approach, because they are operating under different constraints.

All of that said, there is one particular case where "bury it in sand" might be used in a setting that a typical municipal FD might run into. That case would involve a Class D (flammable metals)[1] fire. Those are notoriously hard to extinguish and "bury it" is a valid option. So if a FD had some industrial areas in their district, where a factory was, say, making something out of magnesium or another flammable metal, and somehow managed to catch a load of the stuff on fire, it's entirely possible that the response would be "Call in DOT (or whoever) and have them bring a couple of dump truck loads of sand to the scene". There are specialized fire extinguishers for this stuff[2], but they're expensive, only contain so much extinguishing agent, and aren't - in my experience - commonplace on fire apparatus.

In practice, this situation doesn't come up terribly often. And a plant or factory working with lots of flammable metals in high volume might well have a private industrial fire brigade that specializes in fighting those fires.

[1]: https://en.wikipedia.org/wiki/Fire_class#Metal

[2]: https://www.monroeextinguisher.com/product/ansul-redline-30-...

And it makes me think of fires caused by accidental optical focusing of the sun such as a polished metal dog bowl as one odd example and others such as hanging glass artwork.

We had our windows washed last year and came home to a fresh burn mark on the floor in the hallway from where apparently the sun had been in the exact right spot with clean enough windows to focus a beam of light... We were fortunate that it didn't focus on something more flammable than a hardwood floor.

Or catching a translucent bottle of hand sanitizer sitting out in the console of your car

a buddy of mine is an architect. he wants to design a building with a concave facade such that when the sun is just right it produces basically a death ray. "imagine you're sitting in your office, causally glance out the window, and see a bird explode for seemingly no reason... " heh

The Walkie Talkie building in London has already achieved this: https://en.m.wikipedia.org/wiki/20_Fenchurch_Street#Solar_gl...

So, Cathedral of Christ the Light in Oakland, basically.

At my college appartment we had a fishbowl with glass fish in the window, which at the right time of year focused light just right on the couch. We saw a wisp of smoke one day, but there were several little burn marks in a line from previous days.

I accidentally burned my dinning room table after leaving a magnifying work lamp on the table the night before that caught the sunlight through a window in just the right way to focus it onto the wood.

A local hardware store burned down a while ago because of an oily rag. They had a spill in the paint department, and the employee who cleaned it up wasn't aware of the precautions needed.

You don't hear about this danger as much as you used to--probably because a lot more paints are water-based than used to be the case.

You hear about stuff like that way more than you used to because every time you go on the internet you have to skim over the low effort comments pointing out the rare failure modes in which everything is dangerous to find the real content.

The world doesn't need a lecture on proper jack stand usage every time vehicles are mentioned and doesn't need a lecture on linseed oil fires every time deck finished are mentioned but it gets it anyway.

I for one am pretty happy about the linseed oil stories in this thread because the low ignition temperature is something that I wasn't aware of.

How? There's a warning on literally every product containing linseed oil even in the EU.

I don't have any, never read the packaging, still appreciate the warning because I don't actually read the instructions on every bit of kit that I get, usually they are full of warnings that are either dead obvious ('may cause rash') or ass covering ('may cause cancer when ingested'). Over the years I've become quite numb to these which is a danger in and of itself.

So this is actually a strong motivator to start reading those anyway even for things that seem quite innocent.

When dealing with industrial stuff that I know is dangerous I go over the MSDS in detail and make sure I take all of the proper precautions.

Products have warnings about approximately 100 gazillion things that can happen that basically no one reads.

I think that's true for Americanized consumer products which always seem to come with about 2-3 pages worth of "don't do this" in the manual. For stuff like paints it pretty much just says it on the tin: "Flammable. Harmful if swallowed. If ingested, contact poison control center. Do not induce vomiting."

The longest safety hints I've found in a survey of my paints cabinet were these:

> P102: Don't let kids play with this

> P301+310: IF INGESTED, rinse out of the mouth. Do NOT induce vomiting. Contact doctor immediately.

> EUH066: Repeated contact may cause brittle skin.

> EUH208: Contains lemons, wood oil, cobalt-2+-salts. Can induce allergic reactions.

> EU-VOC-limit (some numbers)

> Safety instructions: Rags and stuff drenched with this should be put in an enclose metal container or under water, since they can self-ignite. Note: product itself does not self-ignite.

If you're familiar with similar products you probably don't need to look at this - but if you use something for the first time, why wouldn't you spend the 10 seconds to understand these, in addition to the prep time you need to use a new type of finish anyway?!

same

Oily rags are super dangerous. Depends on the type of oil or thinner but definitely something to keep in a small sealed metal container.

Linseed oil is the usual culprit. It oxidizes more readily than other oils at room temperature, and a pile of rags can insulate enough for the center to reach autoignition temperatures. The low smoke point for linseed oil also makes it well-suited for seasoning cast-iron cookware (only be sure, please, not to use boiled linseed oil, as it contains lead.)

Linseed oil is just flaxseed oil, and I've found that the latter term is used in cooking contexts, whereas linseed oil is a term used in various industries that use industrial grade flaxseed oil.

I make this distinction because when I went to buy "linseed oil" to cook with on Amazon a while back, it brought me to various linseed oils that were not meant for human consumption.

>low smoke point for linseed oil

I would naively think you would want an oil with a high smoke point for seasoning. Can you elaborate on why a low smoke point is preferable?

My wife uses linseed oil on cast iron. Almost as good as non-stick. Linseed oil has a lot of triply unsaturated α-linolenic acid, which is reactive (due to multiple double bonds) and supports polymerization. A few thin layers of linseed oil added to cast iron and baked to the smoke point create a beautiful sheen and make it easy to clean.

Finally, a reply that is correct. Thank you; too many people leave seasoning up to being some mystery when it's really just that straightforward.

FWIW, I wrote this up a few years back after I went down the rabbit hole of seasoning pans, only to find that it's actually really simple and just as you describe: https://nuxx.net/blog/2014/01/24/well-seasoned-cast-iron-pan...

You want it to smoke when seasoning. That's an indication that free-radicals are being created which set off the polymerization of the oil and create the seasoning. Low smoke-point oils are more convenient because they start polymerizing at lower temps.

You don't want it to smoke when cooking, because ingesting the free-radicals is generally considered unhealthy ;)

The free radicals don't stick around for very long, maybe seconds before they react with something else.

And cooking in general create a bunch of carcinogens like acrylamide, etc. You can reduce it, but it's tough to eliminate it completely.

This is a glaring connection I rarely see people make.

In fact, I've been banging on about it for 15 or so years, and, I kid you not, you're the first person I've seen being this up.

Strange!

Unusual cross section of knowledge: I studied nutritional medicine in a formal capacity for four years 2000 through 2003 inclusive, and I'm also a welder by trade.

At the cross roads of knowledge domains interesting conclusions can be drawn, and sometimes fruitful cross pollination occurs. One thing about the successful founders that I know is that they are not confined to just one domain, and I have been wondering for a long time if that is a trait that helps them to be successful.

Why would boiled oil have more lead?

Lead oxide is a catalyst to polymerization. For wiping down and protecting wooden tool handles it's nice to have the oil harden quicker at room temperatures.

Probably not a good idea to store superglue and cotton together.

Combustion probably unlikely, but non-zero chance.

That reminds me of something. When I was a kid, I dumped a bunch of superglue on a cotton ball once and it didn’t even warm up. I wonder now what the actual conditions are for a superglue cotton fire.

On an industrial scale, sure. For home/hobby use, laying them flat until dry and stiff is plenty sufficient - I treat all such exothermic wood finishes with a great deal of care, but so long as you're not piling up rags or compressing them into a ball you're fine.

Personally I store them in a fireproof container meant to store ashes and inside that they are in ziplock bags. After I am done using them I burn them to get rid of them. Helps that I don’t need stuff like that often.

The primary danger from oily rags is probably from slip and fall on the oil you're working with.

It takes very specific conditions to get oily rags to ignite. As a rule you're gonna have a hell of a time getting any lubricating oil specifically designed not to degrade in a high heat environment because the chemistry that keeps those oils stable over a long time at 200deg (or whatever) makes them very hard to ignite.

A friend left an oily rag in a bucket on a ladder where it was eventually exposed to midday sun; that was enough to ignite it. Fortunately we saw it just as it happened.

Talking about that, lots of lithium pouch thrown in bins. Be warned.

ps: some videos to show how stupid a rag setup goes into fire https://www.youtube.com/results?search_query=self%20combusti...

I had a water heater almost start a fire recently. Installer didn't shield some plastic insulated cable at a point where it exited through a metal hole. Small nick in the insulation plus a little time. It made several 4 inch diameter black spots before the breaker blew.

I was once treating my window ledge with Linseed Oil and was quite surprised when the cloth set on fire from direct sunlight ignited!

An AvE aphorism:

    If your welds sound like bacon you’re doin’ just fine. If they smell like bacon, partner, you’re on fire.

Given he has covered a couple of deadly crane collapses, I'll be surprised if he doesn't end up talking about this accident too.

I had never heard that one. I do have a strong memory of a small bit of spatter from a welding arc ending up in my boot. Never knew I could dance that well. That's why I prefer to never use resin core for wire welding because it spatters like mad. Better use solid core and gas, if it is available (unfortunately not always).

I'll add that if you're not wearing an apron, definitely don't tuck in your shirt!

I hope that you got by the particular bit of knowledge in the theoretical rather than the practical manner.

I do have a strong memory of a small bit of spatter from a welding arc ending up in my boot. Never knew I could dance that well.

I had the same experience during Welding class in high-school. Not fun. :-(

Before I got my auto-darkening mask when first learning GTAW welding I had a few close calls. The window is so dark you simply have no sense that a fire has started until it may be too late, you can't see anything at all unless it's the brightness of a lightning bolt.

Even with the auto-darkening window the field of view is substantially reduced.

I wonder if in the future welding masks will integrate fire alarms.

> I wonder if in the future welding masks will integrate fire alarms.

I don't really see that as a feature anyone wants.

It's pretty easy and straightforward to buy a fire/CO alarm and keep it near you. If you need to be mobile, it seems straight forward to put them on a cheap chest rig.

I get that there might be some convenience factor involved, but:

1. It's pretty terrible UX to add weight to people's heads. Even a little bit makes everything a lot worse.

2. Having an alarm integrated into a mask is going to a Gillette razor blades / HP inkjet situation, where the alarms have to be changed out every 6 months and cost $200. There is great virtue in staying on the path of high volume and multiple suppliers.

I could see it being an additional feature to AR masks, add a ir sensor and some machine learning to classify as welding spark vs material or liquid fire and you don't need to change parts.

May end up being a good use for a VR headset. Pass-through video with HDR support.

This is actually how HDR image processing was invented:

http://wearcam.org/mannventionz/mannglas.htm

So besides 1) looking like a cyberman and 2) likely being dangerously clunky for working up high, I wonder why these didn't take off?

Steve Mann is a super interesting and very nice character.

Oh that Steve Mann.

Yeah I probably fished that idea out of long term memory.

This seems like such a cool idea that I wonder if anyone's already working on it. I've only done the most meager amount of welding, but the "position the tools exactly where you need them to be, snap your head to flip the visor down, do the task while barely being able to see what you're doing, then manually flip the visor back up" workflow seems like it could be improved. The auto-darkening visors mean less fiddling with the mask, but you still can't really see what you're doing, and I think the jury is out on whether or not long-term use of the auto-darkening visors will damage your eyes based on the teensy amount of flash exposure you get with each weld. I wonder if a specialized camera could provide better visibility in the extreme brightness.

The dynamic range for the types of sensors needed would be immense, even with Neutral Density filters...sounds like a fun project!

I’ve seen pictures on reddit of guys using a piece of cardboard and a smartphone as a “welding mask”.

I meant to ask: Do you guys close your eyes for a moment when you start the weld, or just burn your retinas a little bit?

Sounds like the latter...

I believe my machinist friends call this the "safety squint". :P Their attitude appears to be that a little bit of flash from occasional welding isn't going to hurt you; it's the professional welders who're at it all day who need to take the most care.

I’ve heard it as the OSHA squint although I do like the alliteration of your’s.

I close my eyes. And I have a pretty fast set of glasses even so it's not enough to be able to avoid a blind spot after an hour or more of work. Better safe than sorry. The auto darkening glasses make it much easier to start a weld though, you know exactly where you are relative to the workpiece, which especially with stick welding is the difference between a lot of cursing + frustration and some reasonably good work.

Interestingly, I've seen the equivalent of this on the low end - workers in third-world machine shops or electronics recycling places with an opaque block between them and the workpiece, with a cell phone and its camera slotted into a cutout. Just look at the phone screen, not the workpiece.

I consider a CO2 Fire extinguisher as standard as a metal brush. Keep it just as close as your brush and the pin out. I used to TiG and the fire risk is at least predictable and reasonably avoidable. With a MiG, you can start a fire 8 feet away.

The auto-dark can be a mixed bag as you leave it down longer and less likely to notice a fire due to the field of view that you mention.

I'm not sure how you would build a fire alarm. Coatings tend to smoke so detecting ionized air would probably just drive you nuts.

Modern auto-dark shields have periferal vision windows.

One of the most used brands around these parts is SpeedGlas by 3M

https://www.awsi.com.au/3m-speedglas-9100xxi-welding-helmet-...

they will not. ever. wanna know why? economics. until economy is taken out, it will always prevail

That's why we have rules, regulations, and laws.

exactly my point. economy is taken out when you apply a law or a regulation

on what OP idea suggested we currently have market/economy as the one that dictates it, hence none implemented this

Quite. Once I was cutting out the rusted exhaust of a commercial truck with an ox-acetylene torch when the guy two bays away from me decided it was a good time to apply underbody coat to a new truck. The head mechanic started screaming and running around opening bays and we both stopped, probably saving us all.

This is also a reason why many industrial shops (manufacturing or major repair centers) have separate bays for things like that. It's harder to set up for smaller operations, but by keeping them physically separated (never applying paint or flammable things in the vicinity of where a welding torch might be used) you eliminate a great deal of risk.

Crappy electrical appliances (anything that charges a battery, powers a heating element or spins a low torque electric motor) cause far more fires then welding and grinding and other hazardous operations that internet commenters love to clutch pearls over are just so rare by comparison.

For every hour that someone has an arc struck there's a battery sitting on a charger for an order of magnitude more hours.

That's all true. But: this is in the context of people at work, in particular welders, who just caused a gigantic explosion leveling a good chunk of a major city.

Similar accidents have cause the Notre Dame to be damaged multiple large boats to become extremely large paper weights and so on.

Clearly a battery or a heating element or any electric motor (low torque or not isn't all that important really) are also sources of fires. But that's in a different context and typically will not cause disasters of this magnitude due to the difference in context of application.

What with a pretty good overlap between the 'hacker' and 'maker' scene it shouldn't surprise you that on HN people have seen a disproportionate number of fires or almost fires from rather more interesting causes than their electric blankie or battery charger. Though I'm sure that they happened too, none in my life so far.

> It really doesn't take much.

Setting fire to ones own clothes is surprisingly easy with a grinder, and when you’re up a ladder that’s particularly bad.

Oh. That is a recipe for disaster or a rapid improvement in your climbing skills. Ladders are dangerous regardless, combining them with other dangerous tools (chainsaws, for instance) sounds like a multiplication of Morts rather than an addition.

> It really doesn't take much.

I always remember the Imperial Sugar Factory. _Dust_ is enough. Sugar dust is violently explosive.

https://www.youtube.com/watch?v=Jg7mLSG-Yws

Dust is pretty much ideal from a reactive surface point of view, it is even better than a fluid because it readily mixes with air and is immediately explosive in that combination, the activation energy required to set it off is next to nothing.

> Surprise: a box member of a car filled with PU foam.

Groan. I (novice) lit my Xterra on fire while welding on the wheel well. I was going at it and then heard that "whoomph" you never want to hear, and thick white smoke started coming from under the dash. Taking a chance, I ran to grab a small fire extinguisher, pointed it up under the dash, and emptied it. The cabin was full of that nasty powder but the fire stopped, and I never did find what was on fire. All the wiring seemed to work by some miracle. But yeah, that could easily have taken the house with it. Definitely know what's behind the weld.

> when working on a roof a friend of mine did not properly calculate in the effect of an exothermic reaction in a vat of resin exposed to the sun

I see now. The other two were obvious, but roof work seemed counter-intuitive. But if epoxy is used in the process, then yeah, I totally see that. I ended up with a garage full of smoke a couple times from working with more resin than I really needed for my projects.

Retrospectively we came to the conclusion that he had mixed in too much hardener. Incredible how fast that went. From 'hot to the touch' to 'poof'. Nearly colorless flame too, super dangerous. Fortunately I got it covered before it got worse. That wasn't heroic either, I had to get past the bloody vat to get to safety so I figured I'd better stop it burning first.

Roof work sometimes involves applying tarpaper with a.small flame thrower. That genereates quite a few fires.if you are not careful.

It blows my mind that the torch isn't fed with two hoses: One of propane-or-whatever, and one of CO2. If something goes badly, you literally move one finger and start extinguishing the fire.

Why on earth would you want the problem closer at-hand than the solution?

That's simple - cost and convenience. You would have to buy, maintain and lug to the roof yet another heavy preassure vessel. Also twice the hoses & more likely the two will get entangled.

Having a fire extinguisher at hand seems like a better option.

Yup. My uncle's house burned down, years ago, due to careless use of a torch to apply tar roofing.

But presumably also the common use of big torches for materials like torch-on-felt?

I think you're completely correct. The San Diego ship fire basically totaled a amphibious assault ship (helo carrier) was also apparently caused by a welding accident.

Reply on news.ycombinator.com