CA crews handle tricky fire at Tesla factory | Hacker News Reader

CA crews handle tricky fire at Tesla factory

firehouse.com

124 pointsmindcrime96 comments

I'm wondering what process uses molten metal. Or if it was another process that ended up creating molten metal (like a battery fire of some sort).

Given the hydraulic fluid mention I was also speculating that a gigapress might be involved.

Like everyone commenting I am always amazed at exothermic events that can't be put out by water or CO2. The only one I've ever seen "in person" (other than the obligatory set some magnesium on fire in chemistry) was a halon dump on an IBM System 370 when the water coolant system burst a pipe and shorted the many bus bars. That one triggered a 15 second siren followed by a halon dump. I found it quite exciting, the plant manager, not so much.

Giga press.

> That one triggered a 15 second siren followed by a halon dump. I found it quite exciting, the plant manager, not so much.

I am reminded of The Vax War - https://www.hactrn.net/sra/vaxen.html

Some may recall its opening (and closing) line:

> VAXen, my children, just don't belong some places.

The story involves a VAX 750, and an IBM 3090.

The passage that will get you to start shaking you head is:

> The fire alarm klaxon went off and the siren warning of imminent halon gas release was screaming. We started to panic but the data center manager shouted over the din, ``Don't worry, the halon system failed its acceptance test last week. It's disabled and nothing will happen.''

Thank you!!! Was trying to find this a week ago, but couldn't find the right set of search terms.

Is that a true story, and would that really have impacted Black Monday?

I never expected to find an anti-VAXer on HN

I'm pretty sure the machine in question is the giga press.

https://en.wikipedia.org/wiki/Giga_Press

One of the largest pressing machines in the world. I think it currently presses 1/3 of the frame of a model y but they're hoping to press an entire model Y frame eventually.

"press" is a bit of a misnomer, as it actually does die casting.

Die cast? Just like Matchbox!

Thank you for this I was trying to figure out what was being pressed exactly.

What is special about the GigaPress die casting?

Being able to make large, complex parts in seconds to high tolerances. A way cheaper way to make a car if you figure it out.

Injection apparently happens in milliseconds, which is crazy at the scale of half a car.

But you get weaker parts

Not really any more. X-rays and other process improvements have allowed modern die casting to approach hot forged properties. You don’t get grain flow as good as cold forging, but it’s pretty dang good now.

> Being able to make large, complex parts in seconds to high tolerances.

You don't get that high tolerances with casting the bigger you go. Thermal warpage needs correction, and there is only so much you can correct.

But the machine in question is used for a process much closer to injection molding than casting.

MIM is somewhat better with warpage

It's all relative, the basis for comparison is a standard automotive structure: stamped steel parts welded together.

Stamped steel is excellent on manufacturability, warpage, and speed of manufacturing included.

The industry after all spent a century working with it.

Aluminium alloys bring along own problems, but otherwise unbeatable on weight even if very weight inefficient methods are used, like casting.

It's a very complicated compromise even if economics is not involved.

Correct, this isn’t mere casting. It is die casting, with really good surface finish (which matters for strength) and good tolerances once you iteratively take warpage into account.

And a big reason they didn’t use this in the past is you’d get bubbles which reduce strength dramatically. But improvements in process tech and ability to use X-rays to do fast, non-destructive evaluation allows near-forged material properties. Better than castings of years past.

Is this intended to fix the misaligned body panels I constantly see on other Tesla models?

Largest in the world? It's 1/10th the pressure of the Mesta / Alcoa 50,000, which itself is still far from the largest...

https://en.m.wikipedia.org/wiki/Alcoa_50,000_ton_forging_pre...

largest diecast press

Apologies, I meant in the context of consumer automotive.

Tesla shills in here lol

What's the requirement for being a shill? Liking something and talking about it?

It's an injection molding machine actually, not a real forge, or a press.

It's a "molten metal" machine at the Fremont factory, it contains "molten aluminum" and is used to stamp vehicle parts... so yeah, "Giga press".

https://en.wikipedia.org/wiki/Giga_Press

Its aluminum die casting. This is the product that is produced there:

https://cdn.shopify.com/s/files/1/0173/8204/7844/articles/1_...

Not really casting, the process is more close to a metal injection molding/rheocasting.

> I'm wondering what process uses molten metal.

Casting

> Like everyone commenting I am always amazed at exothermic events that can't be put out by water or CO2

Possibly the most extreme being chlorine triflouride, so reactive it's a better oxidiser than oxygen, as a result it can burn sand:

https://blogs.sciencemag.org/pipeline/archives/2008/02/26/sa...

Wondering why Tesla doesn't have fire suppression solutions in place ready for this type of danger. Surely major Class D fire suppression would be a prerequisite for this type of process?

https://www.elitefire.co.uk/help-advice/detect-and-extinguis...

It wasn't the metal that caught fire, it was the hydraulic fluid. You still wouldn't want to use water on that though.

Everyone down thread is commenting like the aluminum caught fire. That's not what the article says:

"A piece of machinery containing molten metal caught fire"

It's really hard to get aluminum hot enough to combust if it's in a big blob. Powder on the other hand...

Hydraulic fluid under pressure is a class B fire than can very quickly escalate. Sand bags aren't going to be doing much in those situations

https://youtu.be/aj8-JlRqoWc

If you have molten metals class D powder suppression is needed. I'm amazed the old NUMI plant doesn't have both ready for action.

Agreed. Class D extinguishers are expensive, but we're talking ~$500 instead of ~$50 for an ABC (obviously depends on size). Both costs are trivial for any manufacturing company, and frankly I'd be shocked if they both weren't required by code.

The only thing that I can think of is perhaps no one noticed until it was too late? And then the employees didn't want to touch it until the fire department showed up. If there was a hydraulic leak, they used the sand bags to stop the spread of the fluid, and then used foam to actually put out the fire? No clue.

Incidentally, this is why a lot of planes use https://en.wikipedia.org/wiki/Skydrol instead of regular mineral-oil-based hydraulic fluid, despite the former being far more irritating for human handling.

Well, maybe they hadn't gotten around to putting it in yet? Supposedly the building was still under construction.

You'd think fire regulations would need to be adhered to before production was allowed to start. Small businesses have endless compliance check ups for this type of thing and big fines if they don't have safety and fire suppression in order

I suspect the machine itself was also under construction. Someone was probably doing tests on it when it caught fire due to some design issue (like the hydraulic hoses touching the pipes where the molten aluminium flows)

Elon may be a visionary, but his disdain for regulations is well documented.

That press in literally in a tent on a parking lot. There’s lots of things that are missing, from what would be considered normal operating conditions for equipment like that.

Silicon Valley spirit of move fast and break things is still strong with them.

Nonsense, its not in a tent. Its a dedicated building to house the press. See it being constructed here:

See: https://cdn.shopify.com/s/files/1/0173/8204/7844/articles/3_...

Good point, not a tent. It’s much closer to a carport.

Based on the photo above - it's a purpose built structure designed to house that exact machine. What makes you think it doesn't meet the machines operating requirements?

Just because the walls weren’t there when the photo was taken doesn’t mean it will never have walls.

Props to Fire Dept. for having some decent response code enforcement - I mean you would never put off a molten metal; it would burn forever, and while you add water to it, it would only expand. AFAIK the only way to put it off is with sand.

> I mean you would never put off a molten metal; it would burn forever

I doubt that, if it burned forever you could use it to heat water and spin a turbine and produce free power. If you meant “forever” as “a long time”, then sure :P

It would eventually cool down after removing the source of heat keeping the metal molten, which are electric resistive heating elements (assuming it works like an aluminum smelter).

It would burn until all the metal has reacted. I think the point is that due to its chemical properties what you're using to try to "extinguish" the fire is actually fueling the reaction so not forever but you're certainly not helping.

Oh, absolutely. Water would exacerbate things (as you said in your sibling comment), I was just being a smartass about the word “forever” :P

I was curious as to the reason why water or CO2 can't be used to put off a metal fire. The TL;DR is that with a metal fire the high temperatures cause the H2O and CO2 molecules to be unstable enough that the metal can attract the oxygen molecules to fuel the ongoing fire rendering them useless as a fire suppressors. H2O is even worse as once the oxygen is split from the water molecule you are left with hydrogen which will combust and explosion might cause the molten fire to spread further. This doesn't happen with wood burning as the attraction to the oxygen molecules from the carbon being burnt is the same or stronger than the attraction from the CO2 or H2O molecules.

Do you think pure nitrogen would work in this situation?

You'd need tons. It's plausible but I don't think it's practical. Also, magnesium and titanium can burn in pure nitrogen, so it's not a panacea.

Probably not. Nitrogen is roughly the same density as air. It would just get dispersed into the atmosphere before it could do much, unless you had insane amounts of it.

Nitrogen wouldn't work in this situation. In fact nitrogen and aluminum will burn, forming AlN. Normally this doesn't happen because of Aluminum readily forms a "rust" layer in the presence of oxygen, and there isn't enough heat. Molten aluminum in a pure nitrogen environment wouldn't have that layer, and there likely would be enough heat.

Out of curiosity, what would happen if you quickly submerged the burning metal into a very large body of water?

Fire on the metal increases then ‘explosion’ the size of which depends on the metal type and quantity obviously https://youtu.be/tj7S_DNFgEU

Also alkali metals is fun

https://youtu.be/I8tOtZKpi04

> very large

It depends on how large, what temperature, and what type of metal. If there's enough latent heat in the quantity of metal to flash boil a significant fraction of the water, you're going to have a bad time. If you're literally dropping it in the ocean on the other hand, then you'll get a brief but impressive geyser and the fire will go out.

(Unless this is an alkali metal we're talking about. In that case you'll just get a "vigorous reaction" of a size proportional to the quantity of metal. Ex: https://www.youtube.com/watch?v=cc6W6SptLvU)

Another big concern is the water turning to steam. It expands 1,000x. 2 cubic centimeters of water expanding to a 2 liter bottle worth of steam. My facility had its own quick response fire unit (production employees who went through fire and rescue training), and I believe the company helped fund a ladder truck for the town of ~600 people which is adjacent to a decent metropolitan area. Fast response is vital to minimize danger to people and damage to the facility.

Factories have been leveled by water hitting molten aluminum. The Hackaday link below does a great job summarizing the dangers.

https://hackaday.com/2020/12/30/water-and-molten-aluminium-i...

Explosion links:

https://youtu.be/sIJmwF1Qs7M

https://youtu.be/jvddrln900Q (last ~20 seconds)

Source: Used to work at a very large aluminum facility.

Aluminum is also used as fuel in solid rocket propellants.

https://en.m.wikipedia.org/wiki/Ammonium_perchlorate_composi...

We did this in HS chem class, with a block of dry ice that we cut in half and hollowed out. Then we coiled up some magnesium ribbon so it would perch in the hole, lit it with a MAPP gas torch, and closed the block. The magnesium continued to burn long after any lingering air in the hollow could've been consumed. After a few minutes the magnesium was gone and we opened the block, and there was black carbon dust in the bottom of the hollow.

And that introduced the conversation about using the right extinguisher for the job!

> H2O is even worse as once the oxygen is split from the water molecule you are left with hydrogen which will combust and explosion might cause the molten fire to spread even farther

Yup, this is why you never douse an oil fire (say in the kitchen) with water. The fire will erupt! Seen it happen live. Fortunately the damage wasn't bad. One third-degree burn.

A kitchen oil fire is a different effect. They're not hot enough to disassociate water molecules. What happens typically is that the water floats the oil, and the fire, remaining on top of the oil, spreads more rapidly.

Another effect is when the fire (and typically, the cooking vessel) are hot enough to vaporize the water. This rapid vaporization spreads oil droplets rapidly, allowing much more of the oil to burn at once. It can appear explosive, even though it's just rapidly-expanding steam spreading fine oil droplets.

I think "rapidly expanding and exothermic" is equivalent to "explosive" at least functionally.

I'd agree that it's explosive. I suspect they were trying to make a distinction between a detonation and a deflagration. A detonation is an explosion with a supersonic shock wave. A deflagration is an explosion without such a shock wave. An oil fire getting water dumped on it is a deflagration, but almost certainly not a detonation. A metal fire getting water dumped on it can result in a detonation. Detonations are vastly more destructive in general than deflagrations.

Exactly that. Thank you. You know better words for it than I do :)

Yeah in the case of water on burning kitchen oil it just makes a big fuel-air bomb if the oil is atomized enough. Kinda like one of these:

https://www.youtube.com/watch?v=i20zvZ-3MMw

Note the initial dispersion of some atomized material followed by a triggered explosion some time later.

Class D fire extinguishers can be used to put out metal fires. They are expensive however and not all fire departments have access to them.

Class D fire extinguishers are used at air bases in my country to put out plane fires since our fighter jets are made out of a magnesium alloy which can burn.

Can't wait for Elon's tweets praising CA public service.

Do companies that do dangerous stuff need to pay additional taxes for fire protection?

It depends, the larger petrochemical plants actually have their own fire departments with specialized equipment and training.

This reminds me of the Chernobyl nuclear reactor meltdown. They put out the fire by dumping boron and sand on the core from helicopters.

https://www.livescience.com/65515-chernobyl-in-modern-times-...

Handling flammable metals fires by burying the fire load in sand is not uncommon. As other people have mentioned, there are special extinguishing agents out there that handle flammable metals (Class D) fires, but they're expensive, and not all fire departments have them. And many departments that don't specialize in industrial response may have a Class D extinguisher or two, but nowhere near enough extinguishing agent to handle a large fire.

OTOH, large industrial plants often have their on on-site fire brigades that have specialized equipment and training specifically around the unique hazards found at their facilities. Not sure what the situation was here at the Tesla plant. But I can say that if my old volunteer department in a mostly rural part of NC had ever encountered a large flammable metals fire (the most likely scenario would probably have involved something in transport, and a motor vehicle collision), we would almost certainly have started requesting dump truck loads of sand, and a front-end loader.

This sounds like the perfect opportunity for a Derek Lowe link to Sand Won't Save You This Time: https://blogs.sciencemag.org/pipeline/archives/2008/02/26/sa...

Be warned: the "Things I Won't Work With" category is entertaining and can easily take up an evening of your time.

> Be warned: the "Things I Won't Work With" category is entertaining and can easily take up an evening of your time.

Don't forget the "Things I’m Glad I Don’t Do" category with such gems as "There’s Toxicity, And There’s Toxicity." Not as many posts as "Things I Won't Work With", but enough amusement for few good reads.

When you run out of those, move on to Ignition!: An informal history of liquid rocket propellants by John D. Clark.

https://archive.org/details/ignition_201612/mode/2up

I recommend stopping after one or two posts, because he uses the same jokes every time and you should let yourself forget them so you don't notice.

Brilliant article, thanks for bringing it to my knowledge, friend.

Would the Texas FD be prepared for this in the area of their new plant?

My understanding from talking with firefighters is large industrial plants, major apartment complexes etc give the local fire department tours and discuss special circumstances of their building.

even tiny little light industrial shops have an ongoing conversation and yearly inspection with the fire department about doors, locks, stairs, extinguishers, chemicals, flammables, aisles, accessibility, sprinkler coverage, signage, bottled gas storage, electrical, etc. etc.

as a solo operator its a huge pita. I do appreciate that these people are putting their lives at risk, so I suck it up (don't have much of a choice actually)

big of you.

You do understand that this PITA is to offset a larger PITA right? It's a good economic choice.

maximizing cost externalization is the name of the game. smh.

This is correct.

Departments will develop “pre-plans” for any large buildings, industrial buildings, etc.

Yeah it's not like Texas has ever had a factory before.

True, you can’t have a factory explosion [1] without factories.

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

Politically Texas attempts to avoid regulating industries. I don't believe that it is unreasonable to consider if the area in which the factory is being built may not have the strongest requirements and preparations by the local emergency services.

What's suspect is why were they building cars in a building that was still under construction?

«Move fast and break things»?

It seems like the die-stamping machine is outside actually.

https://electrek.co/2020/09/09/tesla-deploying-second-giga-p...

I don't think the die-casting machine is "under construction". It seems like they intend to leave it out there.

I remember seeing some news when it was operational a few weeks ago.

Some footage of it in action on Jan 21: https://youtu.be/BwoiFC-HwPE?t=307

Feb 8: https://youtu.be/hF_oRMIY9tk?t=363

Now imagine a parking garage full of EVs, where one catches fire.

Battery fires are notoriously difficult to extinguish.

Are you saying ICE cars aren't at risk of incidental fire from other [burning] cars, what with their 10+ gallon gasoline tanks?

Are you saying ICE cars aren't at risk of incidental fire from other [burning] cars, what with their 10+ gallon gasoline tanks?

They definitely are (source: former firefighter here), but a deck full of Teslas's is more of a problem for the FD in some regards, than a corresponding deck full of ICE cars. The reason being, Teslas are known to be hard to extinguish when they burn, and are apparently prone to rekindles. I don't remember the exact number, but I seem to recall hearing that Tesla recommend flowing water on a burning Tesla for somewhere around 10 hours in order to cool it sufficiently to avoid a rekindle.

An average car fire (depending on certain details) can be taken care of in somewhere around an hour, from start to finish. So a bunch of Teslas on fire would require more resources committed for more time, which would possibly impinge on the ability of the FD to handle other calls.

Edit: here's the Tesla Emergency Response Guide[1].

According to the "Firefighting" section:

Battery fires can take up to 24 hours to extinguish. Consider allowing the battery to burn while protecting exposures.

[1]: https://www.tesla.com/sites/default/files/downloads/2016_Mod...

Gasoline and diesel (which is harder to ignite) need a source of ignition to start burning, whereas a lithium ion battery can supply its own. The latter being a solid also means heat can build up deep inside a damaged battery pack enough to cause reignition.

This has literally has nothing to do with batteries.

I see I’m not the only one who jumped straight to “lithium fire” upon reading the headline.

Reply on news.ycombinator.com