Hacker News new | past | comments | ask | show | jobs | submit login
Under Pressure (minusbat.livejournal.com)
494 points by FatalLogic on Apr 27, 2015 | hide | past | web | favorite | 96 comments

I do explosions with liquid nitrogen almost daily at my job as a science demonstrator. We've had one accident where a bottle exploded in the hands of a college of mine. He got a good scar up his arm, but nothing deadly. He had to get sewn, but that was about it.

Most often though, if you fill up the bottle with too much nitrogen, it won't explode. It'll just freeze up and slowly disperse through whatever cracks are present. If it didn't explode over night, it probably wouldn't have at all.

I suspect it might have started to fizzle due to air leaving the bag, or due to a small leak, which is pretty harmless. Nitrogen explosions of half-litre bottles are very large, and can shake the ground 20-30m away. I can test it later today if time permits and upload a video, but overall, it would've been much clearer if it went off.

Edit: to people saying the cap is the weak point: 9/10 times it's the bottom that gives in first.

I have some experience with dry ice in closed containers.

Every now and then some friends and I will get some dry ice, a bunch of plastic bottles and other containers with a screw-on lid and have a laugh making small explosions.

Basically you take a bottle, fill it 1/4 with water and 1/2 with dry ice and screw the cap on. You then have around a minute before it explodes.

A 1/2 liter coke bottle will make a nice whooompf and a 20 liter plastic gasoline can (the largest we have tried) will make a bang that can be heard maybe half a mile away. They very rarely sizzle out, and the ones that do probably have a defect, or we didn't get the cap screwed on properly. There's some time pressure, so you don't stand around checking before you throw the container. I would imagine that a metal container will make a pretty big bang (higher pressure before it ruptures) and may throw out some nasty debris.

It's good fun!

Back when I did this a few times a year, 20oz soda bottles made the best boom. 2-liter bottles couldn't take as many PSI and water bottles... well, they're not designed to contain pressure. They'd fail at the cap, usually.

That may no longer hold, since 20oz bottles feel flimsier these days and all have low-profile caps. Haven't tried in a while.

Also, PVC pipe burried in ground, drop dry ice bomb in, drop another bottle with some water in it on top = dry ice mortar. Those little plastic bubble things that toys in vending machines come in? A little dry ice, a little water, close it, place lid down. POP, the bubble part flies a meter or so in the air.

I will note for anyone trying this that the parent's ratios are very different from what I used. Crushed dry ice to 1/10-1/8 full, about twice that much water. Unusually warm water (say, from near the surface of a pond or lake in late August) will greatly reduce time-to-boom, so beware. Too little water and it'll freeze before boom, greatly delaying or even preventing it. Very annoying. Attaching to something heavy (but NOT shrapnel-genrating) and sinking in ~5-10 feet of water is fun. Huge bubble, explosion can be felt on land nearby.

I used to do this too when I was a kid at my grandmom's house. They'd order steaks from TV commercials and receive them in the mail bundled with dry ice. It started as making foam messes with soap, water and dry ice... But soon I was filling up 2 litre bottles with ice and water and making bombs. It all ended when the police arrived because neighbors thought a gunfight was happening in their quiet residential neighborhood. Needless to say, they weren't very happy with me.

Big difference, I think, is that in your case you're mixing it with water, so it will warm up a lot more quickly. In the OP's case, he just had the dry ice in there with no water, so presumably it would melt more slowly, and there may be enough time for the carbon dioxide to escape through tiny gaps or cracks in the cap.

> to people saying the cap is the weak point: 9/10 times it's the bottom that gives in first.

It may be the case for a typical water bottle but thermoses have much larger caps. The maximal force that a cap can hold is proportional to its perimeter (~r), the force itself is proportional to the area of the cap (~r^2) and the pressure. So larger caps can blow off by lower pressure. I would bet that for a thermos the weak point is the cap.

Do the amount of splines and their distances away from each other have anything to do with it?

> I can test it later today if time permits and upload a video

+1 - I'm always up for a good explosion video!


Sewn, i.e. stiches in his arm.

I'm more curious about this animate college :)

Paraphrasing from that article: "So, realising I had created a bomb, I surrounded the bomb in shrapnel and took it to a crowded part of the city".

As soon as I read the bit about putting it in a bag packed with tile I was like, "this doesn't end well".

That was a pretty cool read.

Yeah, that guy should be arrested or at least given a heavy fine :/

"I had to keep zig-zagging to avoid pointing the bit which was going to explode at people coming up the street towards me" and "the canal is a crowded place on a Sunday morning"

My god, man :/

Why? Nothing bad happened, so what is the point of punishing him? He learned something and went home. So did everyone else.

Endangerment is legitimately considered a crime.[0]

Our go-to example in an Economics of Law course was firing a gun while in a crowd. Similarly, attempted murder is a crime even in the case where no harm befalls the victim.[1]



The value of endangerment laws to society, to the extent that there is any value, is to deter the offender and deter others as well. In the story here, I don't see how you can apply that ethic. I don't want to live in the society in which every low-probability danger is made into an offense. Life has some sharp edges.

Luckily there's a common sense rule to apply that an economic analysis of law makes clear.

Enforcement should occur up to the point where the marginal cost of extra enforcement is equal to the marginal cost of the activity we are seeking to deter.

I would agree with you that the particular circumstance described in the article does not warrant legal action, but your original post made no allowance for enforcement directed at anything other than purposeful and effective crime.

That logic doesn't seem to apply when a cop pulls you over (while still intact) when caught driving 200km/h on a highway which is not the Autobahn.

"Why? Nothing bad happened,"

duty of care?

A few calculations about the possible explosion:

- Assuming the inside of the thermos is at room temperature (and that there was enough dry ice initially), the pressure should be around 60bar. This is the vapor pressure of carbon dioxide at room temperature [1].

- At 60 bar and with a volume of, say, 1 liter, the energy available for the explosion is roughly 60bar x 1liter = 6kJ [2]. This is a TNT-equivalent [3] of about 1.5g, or about 10-100 firecrackers. Enough to cause injuries, but not enough for structural damage to a balcony [citation needed].

In my personal opinion, the most dangerous thing was handling the thermos. I believe letting the thermos sit on the balcony for a few days and closing doors and curtains (to prevent glas shards flying in) would have been a much safer alternative.

[1] http://en.wikipedia.org/wiki/Carbon_dioxide_data#Vapor_press...

[2] https://www.google.com/search?q=60+bar+*+1+liter&ie=utf-8&oe...

[3] http://en.wikipedia.org/wiki/TNT_equivalent

Your energy estimation makes no sense whatsoever. The figure you gave is for isobaric expansion, which is an entirely different beastie.

Also, he mentions ~100bar, not ~60 bar.

When I did a first approximation assuming adiabatic expansion, I got ~22KJ.

See https://news.ycombinator.com/item?id=9449086

You are absolutely right, I should have used the formula for adiabatic expansion.

However, I maintain that the pressure inside the thermos should only be 60atm, since this is the vapor pressure of CO2 at room temperature [1], where the liquid and the gas phase are in equilibrium. This is like the butane in a lighter: Butane evaporates at room temperature, but there is an equilibrium between liquid phase and gaseous phase when the pressure is higher.

Using 60bar and the adiabatic formula, I get 12.5kJ of energy.

[1]: http://en.wikipedia.org/wiki/Carbon_dioxide#Physical_propert...

Ok, that works.

I was just going from the article's figure of 100atm. I probably should have double-checked that figure.

Great read, but speaking as someone who lives on one of those houseboats (and passes through Islington occasionally), please don't consider a canal a safe place to 'ditch' potentially exploding things in future.

"I looked at the flask with that awful sinking feeling you get when you realise you have created something which is inevitably going to explode at some point in the future, and there's nothing you can do about it."

A feeling we have all experienced many times before, no doubt.

Probably the same feeling that when you lie to someone, see your lies being slowly exposed, and there's nothing you can do about it. Happens a lot when we are teenagers :)

Why not just put it outside on the balcony and leave it the hell alone? Fussing about it any more is just risking getting your hands blown off. Idiot. And all this faffing around with a time bomb while you have a kid to look after, totally irresponsible. Put a duvet over it and leave it alone. It's nearly Darwin award stuff

Seriously, why not just wrap it in blankets, put it all into some open lid container (trash bin, hamper, etc), and leave it on your balcony. There are so many better ways to handle this situation.

I agree that handling this was dangerous, but I wonder how much damage the explosion would have caused if he hadn't thrown it in the river...

In picture: https://www.youtube.com/watch?v=Ldgp3Ton7R4

experiment start at 3:00

Can someone explain how the entire bin went up? Does it mean that the explosion poked a hole at the bottom of the bin?

I expect that the floor of the bin is elastic in the technical sense - it returns to its resting form when no force is applied. When the BANG goes off the base of the bin is compressed into the floor, hard. This lowers its center of gravity. Then it returns to its previous shape, the CoG rising as it does so. Once it has returned to its original position it discovers, rather to its surprise, that it's now travelling upwards, and so it continues to do so.

You also have less pressure inside the bin than outside.

I think the shock wave travels through the bottom of the bin, bounces off the hard concrete floor and back to hit the bucket, thereby lifting it up.

As well as the shockwave mentioned elsewhere, you also have pressure effects.

The initial explosion drops the pressure in the bin below atmospheric pressure. Assuming there's enough kick to separate the bin from the floor, you've got a higher pressure pressing up on the bottom from below than down from inside the container, so the bin accelerates up.

I'm not sure if it'd be a significant effect, however.

A few years ago while on vacation with my extended family, all my brothers and I decided it would be fun to drop dime size chunks of dry ice into a water bottle with a few inches of water in them and then throw them into the pool. We did this and given a few minutes they would explode, harmlessly - not much more than a fire cracker.

Then we had the bright idea of tying a 5lb weight to the water bottle and let it sink. This was a bad idea on our part. When it exploded, it sent a shock wave through the ground. It was pretty intense make us brothers kinda freak out. People rushed from inside the house and asked what explosion was. We were afraid we had cracked the pool, the concussion wave was so strong. Luckily, we hadn't for my brother's sake. It just makes me think twice before containing dry ice.

Can anyone explain why putting a plastic ziplock or other bag around the thermos and hiding inside a cooler (or close-able container box) wouldn't be an adequate solution?

I imagine this isn't some pipe-bomb - the energy density simply isn't there - but you want to ensure that the thermos lid doesn't happen to injure some passer-by.

The amount of energy in a thermos like that is high enough that I'm not sure that would be safe.

Quick approximation, assuming adiabatic expansion:

Gamma for CO2 is ~1.29.

    Pf*Vf^y = Pi*Vi^y
    Vf = (Pi / Pf) ^ (1/y) * Vi
    Vf = (100atm / 1atm) ^ (1/1.29) * 1L
    Vf = 35.51L
    W = (PfVf - PiVi) / (y - 1)
    W = (1atm*35.51L - 100atm*1L ) / (1.29 - 1)
    W = 22.31 KJ
For comparison, a .50 cal round has ~15kJ of muzzle energy. Now, a lot of that energy won't be focused (if nothing else, the final temperature with adiabatic expansion is such that a large chunk should sublimate again), but still.

The 22 KJ seems widely varying form other estimates on this discussion (6KJ, 15KJ).

For example, is it clear that the pressure of 100 atm would actually be attained by a simple thermos and dry ice? What lead you to use that number?

Furthermore, the .50cal round is focused on a small area, the impulse and destructive force are multiplied by the shape of the bullet. Just like how shape-charges magnify the explosive force of munitions to bust armor.

The 6KJ number is flat-out wrong. You cannot multiply pressure * volume like that. It's not an isobaric expansion. I mentioned this in a reply to his comment.

As for the 15JK number, I cannot see any estimates in this thread saying 15JK. Could you link the comment?

The 100atm figure is assuming that the linked article's calculation of final pressure is correct, assuming the thermos doesn't burst beforehand. Although I fully agree that a standard thermos is unlikely to achieve that number.

And as I said a lot of the energy won't be focused. This is just a first approximation, to indicate that yes, potentially the energy is there.

Edit, because YC locks comments for some absurd reason:

Apparently the vapor pressure of CO2 is ~60atm, which bumps the number down to ~12.5KJ. Though I haven't checked that number yet.

I have to ask - can you explain the equations please?

They all come from the equations of adiabatic processes of an ideal gas. Wikipedia explains better than me:



(Especially the section "Derivation of discrete formula" in the latter)

It's only a first approximation, but meh.

Curious: how much damage could this have done if it actually exploded? What kind of forces are we talking about here?

Not much. At the student revue last year, we had filled a thermos with dry ice, closed it up, and forgot about it on stage overnight. At some point during the night, it blew off the plastic valve at the top, which was ruined, but the metal flask itself was fine. The people sleeping near the stage reported a loud bang, but nothing else.

Hypothesis: since the valve is only going to be propelled by the expanding gases for a relatively short time before it has cleared the flask opening, I don't even think it accelerates to that great a speed.

Clarification: a thermos is more or less a best-case scenario, since the valve at the top is going to be much weaker than the flask itself. A plastic bottle is worst-case, because it is going to shatter and eject plastic splinters.

Could have seriously damaged his hand if he was holding it. I have a permanent scar in my eyebrow from dry ice enclosed in a weak water bottle that exploded a few (4 or so?) feet from my face. There are YouTube videos of people ending up with severe hand damage from 2 liter bottles.

I was cringing reading this article. I would never ever have handled one of these things, based on my experience. I think I would've put it in the fridge, left the house for at least a day, and bought a new fridge based on the damage.

an aluminum scuba tank is filled to around 3k psi; they're quite energetic if they explode


This guy died and damaged cars 100ft away from the center:


Now, the article says 100psi not 3k, but you'd still probably rather not be near it.

100 atmospheres is approx 1500 psi.

I inflate bike tyres to 110 and its nothing tbh, 1500psi is a different kettle of fish.

Wow. In the first example, it seems those weren't scuba tanks, but rather pressurized air containers that are used to fill the real scuba tanks - they are quite a bit larger. Never seen those before.

Actually the article says that people are also killed by exploding truck tires which are around 100 psi and further states that SCUBA tanks are pressurized at up to 3000 psi.

Isn't the pressure only part of the danger? In my non-scientific experience, it seems that the speed of failure is the important part. i.e a truck tire with a nail in it does no harm, but a 15-ply semi tire at 100psi having a catastrophic failure is a much different animal. All about the speed of release.

My own personal story: some friends and I filled a 5 gallon bottle with a bit of isopropyl alcohol, shook it up, and lit the opening to make a "rocket flame". We had seen this done in class, and it went fine. BUT...we wanted to do it again. we had no more alcohol, so we used Acetone instead, but it wouldn't light. There wasn't enough air in the bottle, since we had just burned out the oxygen. Because we were 16, and lacking much foresight, we decided "why just put air in the bottle, when we could use pure oxygen from a welding tank?". We did that. The ensuing incandescent explosion lit up briefly like a lightbulb and then ruptured the bottle into about 100 pieces. the major one landed 2-3 seconds later, about 200 feet away.

Even though that was moronic, I pride myself for having worn welding gloves, a face shield, ear plugs, and used a 12 foot handle with a match on the end. The detonation left me feeling shaky and jittery for about 6 hours.

My point is, speed of failure is important.

That story belongs over on the "Under Pressure" thread! - https://news.ycombinator.com/item?id=9444675

I think it is?

Well I personally know a physics teacher who does extreme physical experiments using dry ice and liquid nitrogen. One simple experiment is pouring really little amount of liquid nitrogen into a 0.5l coke bottle then wait. We stood away from the bottle (~10m) and let it explode. It is really loud when it bangs, otherwise it's a safe experiment (of course it's done outside and you really shouldn't hold it in your hand).

What's the weak point of a thermos? It's definitely the cap. I think leaving it alone would have just sent the cap flying, it could have been safely done in a park.

That there was a continuous stream of small bubbles (as opposed to one gigantic one) suggests that the failure mode was a leak, not an explosion. Still, I would have been scared.

The odd part is that submerging in a depth of water actually lessens the pressure differential making it less likely to explode, not more. So failure is less likely to occur as it's submerged further.

I'm not sure. The stream of bubbles could have been the remaining dry ice sublimating, couldn't it ?

thrilling read; interesting that the device manufacturers explicitly warn you against putting dry-ice in a thermos. I had never heard that before.. I mean, thinking about it- it makes sense, but I would not have reached the conclusion without assistance.

When I bought dry ice (last year) for the science fair, they do give out a safety brochure on handling.

One of the top 5 points, no tightly closing containers. Styrofoam cooler is ok and dry ice will last 2-3 days like that.

Pete is the sort of guy who needs a Skippy's List. See his LJ icon? That's a burning NextCube case.

My instinct would have been to cool it in some liquid nitrogen at the university and just safely let the dry ice out... Glad they had a canal nearby...

Why not just puncture the thermos?

He was working on the assumption that there was significant pressure built up in the thermos. Puncturing it would release that pressure causing a potentially dangerous explosion. The same premise as puncturing a balloon except the balloon in this case is made of metal and glass and the amount of pressure is significantly higher.

If you're lucky then the top of the thermos is plastic.

Safe way to solve the situation with a metal thermos with a plastic cap: clamp thermos to workbench, pre-drill hole in large sheet of plywood, drill through pre-drilled hole into thermos. Depending on how complete the sublimation is, there may be a boom and you, the plywood, and your drill may potentially get kicked back a meter or two in the worst case, but your plywood shield would likely spare you from any high-velocity plastic fragments that would cause you injury. If you're lucky the pressure is still low enough that it just blows plastic out of the spiral of the drill bit and there is little other than the plastic shavings to work around.

You would definitely want to be wearing earmuffs though.

Funny. Being from a very rural area my solution would of been to walk outside and hoof the bloody thing as far as I could into the surrounding fields and have a cup of tae while waiting for the boom.

Being from a very rural area of the US my solution would've been to leave it in a field and shoot it.

Reminds off the stories of unattended luggage getting shot by security personnel at airports [1].

Aparrently, shooting is the preferred way to disarm potential explosives in such scenarios.

[1] http://lilyasussman.com/2009/11/30/im-sorry-but-we-blew-up-y...

It's a good idea to own a few basic tools, including a handheld drill and maybe a hacksaw, even if you live in a small flat.

Although in this case, depending on how long the thermos had been shut, I'm not sure whether attempting to relieve the pressure would have been the smartest move.

Now I'm genuinely wondering what I would have done...

Like dsfsdfd said, leaving it on the balcony under a heavy blanket seems like the best option, though my first thought was just throwing it down the street sewers.

Yeah, thinking about it some more, "wondering what I would have done" quickly turned into "wondering whether I'd still have hands or a face right now." Scary stuff.

Drilling a hole is not a good option. If there is significant pressure inside then the hole will begin the explosion, not release the pressure. In an ideal world, the best option is to puncture it from far away, which is difficult out in a city. In the country, you might find someone to shoot the thermos with a small rifle.

An explosion in a drain? Surefire way to cause a lot of shit....

My college roommate had a volumetric flask with water and dry ice explode in his hand as he held his thumb tightly over the top. No injuries, and we laughed at him for being an idiot.

So the solution was the place the "bomb" in the public canal where it could hurt some poor swimmer or small boat when it eventually ruptures.

Seems very irresponsible to me.

I don't think anybody swims in the canal in Islington.

I've never seen anyone swimming in it - plenty of kayakers and narrow boats though.

It was the best solution he could think of at the time. That's not irresponsible, just human.

> That's not irresponsible, just human.

The two aren't mutually exclusive.

Funniest story I've read in a while :)

As long as he didn't put aluminum and Drano in the bottle..

The lesson of the story? Don't have kids.

wow, i listening Under Pressure by Queen while writing this comment. :)

This seems to be crazy. Invasion of privacy.

Please add a more descriptive title. Is that some kind of autobiographic short story? Is it something that teaches how to be a better entrepreneur/developer/designer? What kind of people would be interested in reading that?

edit why is it so hard to say that there is a novel to read and not some specific information? It's okay to be a novel, but some people don't want to read novels.

I submitted this. I would have preferred to use a more descriptive title, but HN guidelines discourage changing titles.

"Otherwise please use the original title, unless it is misleading or linkbait."


It's not misleading, I think, just uninformative.

That's good to know! I would interpret it differently, though.

The idea, as far as I see it, is that you don't use title changes to increase the attention, e.g., avoiding to use upper case. But you should certainly use the title to increase the information density, e.g. 'translate "10 Ways To Do X" to "How To Do X,"'.

But I see your point.

Sometimes people just write about their experiences, and it's your job to work out what lessons to take away from it, not theirs.

Sometimes you have to think about which articles you open to read and the title should tell you.

Or you could spend a few seconds skimming through and see if it interests you. If you're that busy you probably shouldn't be on HN to begin with.

You are right, and I basically do that and mostly write that kind of comment only to threads where I don't have an idea if I want to read it or not, even after skimming through (the first page) of that kind of article.

Look. No matter how wrong or stupid I act, If 10k people watch your link and everyone spends a second considering if they should click it or not, then that's a lot more time compared to choosing a title that says more about what to expect after the click, right? So even if I'm the biggest douche on the planet, we might still agree that it's worth for a link poster to spend some time about its title instead of just copy&pasting the article title.

Fair enough, I agree.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact