Tesla big battery fire under control after burning more than three days | Hacker News Reader

Tesla big battery fire under control after burning more than three days

theguardian.com

94 pointsmikro2nd81 comments

So, it seems a little misleading to describe it as a "Tesla" fire:

"The 300MW battery project is being produced by French renewable energy giant Neoen..."

I mean, yes, Tesla made the batteries, but they were not the ones who were in charge of this installation. It's a little like describing a home fire by the name of the lumber company whose lumber is burning.

This is funny. If it was a successful installation, we would see headlines like "Tesla 300MW battery makes X millions a day". Now that there is problem, you want to shift away from Tesla?

https://www.teslarati.com/tesla-megapack-battery-farm-ground...

https://newatlas.com/victoria-big-battery-tesla-300mw/

etc.

You're implying that the people shifting positive stories towards and negative stories away from Tesla are the same people. You might be right, but that's one hell of an assumption.

More precisely, Panasonic made the batteries. Tesla assembled those batteries into packs. Neoen assembled those packs into an installation. And those are just the big headline companies we know about. For all we know, an independent forklift driver might have accidentally dropped that particular pack during shipping and didn't tell anyone.

Panasonic is the licensee of the factory while Tesla is licensing the technology for manufacture.

Tesla is the high profile company here. It seems normal that if they get the big headlines for positive events rather than Neoen or Panasonic, the bad events are likely going to fall on them too. In this situation, the high profile company is incentivized to enforce high standards across the industry.

The alternative (getting all the positive PR to well-known companies and blaming lesser known companies for failures in a vertical industry) would create negative externalities.

I doubt there is any prospect of Tesla "enforcing high standards" for any third party company installing their batteries. Can you imagine a lumber company enforcing high standards in how their lumber is used? Or steel companies enforcing standards on how skyscrapers are made? Or server companies enforcing standards on how server farms get made? We'd call that a massive overreach and power grab, and rightly so. This is the kind of thing that led to the "right to repair" movement.

Lumber and steel are comodities. That means people don't really care about where they get their lumber/steel from; they only care about specific physical properties of the material they buy. Companies producing comodities are typically not high profile.

High profile company status typically comes hand in hand with a modicum of influence which can be used to craft restrictive contracts, or to avoid association with garbage-tier industrial partners.

Agree. It's the Guardian's headline, only very slightly changed by me to meet HN wordcount limit.

Sorry, I meant to criticize the Guardian's headline, not you, sorry if it sounded otherwise.

No worries - didn't take it that way ;)

Plus, this is a brand new site, not yet online. Installation may have been a factor or cause of the fire.

It's a preview of what other fires with this Tesla hardware will look like, no matter who installs them.

And that preview is that they're not that bad. No deaths, no injuries as far as I can tell, and only 3 days to get under control. You can easily think of dozens of more deadly and environmentally damaging disasters associated with other energy sources. Solar/wind+batteries seem pretty appealing so far.

And to add to the "not that bad", it seems that the fire was quite contained with only one group burning and not propagating to the whole installation.

There has been numerous cases of battery fires with phones, cars, ... it's not uncommon.

This installation seems to have been planned quite well to prevent a massive disaster.

This is not so much a Tesla hardware issue as it is a lithium ion issue. Tesla's packs are not more or less dangerous than other lithium ion battery packs.

This is not true. There are different chemistries. Tesla hardware, as many others, requires an active cooling system because of the KNOWN thermal issues with involving cobalt. There are lithium ion batteries without cobalt, such as Simpliphi, that do not have any runaway heat issues and therefore do not require a any cooling system, but this lithium chemistry increases the cost.

Active cooling in lithium (or any chemistry) batteries exists to allow fast charging and discharging of the battery, as those actions produce heat. Without any cooling system, you're limited by the passive heat dissipation of the battery, which largely depends on the internal resistance. You can lower the internal resistance at the expensive of some other properties (mostly efficiency).

If you don't believe this is a problem in any battery chemistry, I challenge you to find a battery that does not heat up while being charged (they all do, due to thermodynamic inefficiencies). Thermal runaway in Lithium-ion specifically is largely an effect of the lithium decomposing as a result of large heat input (by a short or physical damage). Once this has occured the battery will go through several stages of burning.

The runaway heat issues, with the chemistries applied now, are mainly due to which other elements are involved. Cobalt is a huge factor.

It is true all batteries heat when charging, but this is not what I am talking about here. Simpliphi batteries do not require any cooling system because the chemistry is more thermally stable and does not have a risk of thermal runaway, even charging at high power. Tesla batteries have a cooling system not for the heat generated by charging alone, but because of the runaway thermal reactions involving cobalt.

I’d imagine lessons were learned from the incident though and the next time it happens, responders will be more prepared ?

That's fair, but also it feels like fear mongering by omission, because whatever happens at a battery plant, it's not nearly as bad as the fossil status quo. How many refinery explosions, oil spills, pipeline leaks, tanker fires, drilling mishaps etc etc have we dealt with? They all demonstrate uncontained release of chemicals and energy is ... bad. Batteries are less bad.

I think your analogy is a little off. It's a little more like describing a car fire (Chevy Bolt) by the name of the battery cell manufacturer (LG Chem) whose battery is burning. There's a lot more potential energy in battery cells than in lumber.

Burning wood has a energy density of roughly 10 MJ/kg. Lithium batteries have about 0.875 MJ/kg of electric energy. Lithium metal burns at about 43.1 MJ/kg. So yeah. By a factor of 4 at least.

That's very generous of you. Another way to frame it is as an Air Crash investigation where Tesla is Boeing and Neoen is the Airline.

$TSLAQ is on it -

https://twitter.com/tweetmyinbox/status/1422153281464328192

Is $TSLAQ still trying to make money or are they now a complex subculture?

That meme needs to die.

Elon Musk won the war with the short sellers once TSLA was added to the S&P 500. That was a game changer.

Now there is daily institutional money flow into the stock. Now TSLA has a much greater probability of survival.

Michael Burry of ‘The Big Short’ reveals a $530 million bet against Tesla (May 17 2021)

Tesla short-sellers lost $38 billion throughout the automaker's colossal 2020 rally (Jan. 1, 2021)

This seems complex.

All anyone shorting has to do is create a drop over a small period and they can make a lot of money. This has been done successfully before using social media.

If by 'That meme needs to die' you mean it's not healthy for a market at their level then I agree. They are using an army of people who are not very smart, also not good.

But their plan at a base level could work and they could get rich. Create a panic for a small period and get rich.

> Michael Burry of ‘The Big Short’

His foresight around mortgage backed securities is legendary. A genuinely brilliant mind with respect to financial instruments. If he ever points to another financial instrument and says it's fishy, you'd do well to listen to him. But this doesn't make one an expert in everything.

I wouldn't expect Burry's predictions of Tesla's future prospects to be any more reliable than Linus Torvalds predicting the future prospects of some novel financial derivative.

The meme that needs to die is referring to Tesla as $TSLAQ. The trailing Q indicates a company in bankruptcy.

The addition of Tesla to the S&P 500 means that bankruptcy is now much less likely than previously.

Huge amounts of money specifically track the S&P, and huge additional amounts of money "closet index". Index money flows into the stock market on a daily basis and creates price support for all S&P 500 stocks, including Tesla.

Create a panic for a small period and get rich.

I agree this is a viable strategy for betting against many stocks. But now Tesla has probably gotten too big for that strategy. I've tried to read Reddit WallStreetBets but there's so much noise. Smarter people than I undoubtedly can make money on the back of that army of "not very smart" people.

Also something not really making big headlines is a oil well fire in North Dakota thats going on a week burning:

https://www.insurancejournal.com/news/midwest/2021/07/29/624...

What-about-ism is one of the worst trends of recent years IMO.

If i can remember the name there is an ancient greek play that laments this phenomenon. 1000BC or so.

Recent years? Whataboutery has been around a very long time, it was particularly associated with the Soviet Union.

Sure, Russian propagandists did it. But you wouldn't necessarily see fellow forum members partake in it until recent American political elites started doing it themselves.

Seems like What-about-ism all the way down around here.

Sure. We can start with you explaining why an oil fire in North Dakota is related to a electrical fire in Australia. Are the policies involved in the creation of the oil well related to the Australian government?

Was there a conflict in funding of some kind? Did the American Congress decide to withhold funding from one project in order to build the other? Are the two projects related in any way?

EDIT: Do people even use oil as an energy storage mechanism? I don't think so. The closest "dirty" energy analog is in fact, natural gas (which are used in peaker plants). I don't believe oil is used outside of baseload power plants.

Look, I'm well read and knowledgeable of many facts and disasters. I can even one-up you, and point out the Centralia mine fire, which has devastated a town and continues to burn today (In case you want to 'PROVE' that I'm anti-greenhouse gas and pro-environment, I know of plenty of anti-coal / anti-oil arguments myself). But there's almost no relevance from the Centralia mine fire vs this Australian battery fire. I'm not going to be bringing that up unless there's some kind of connection.

----------

The classic example of "Whataboutism" is when the Russians are asked about gulags, they respond about riots in the USA. You ignore the original question and instead distract the issue by talking about something your rivals are doing.

Just puts into perspective the news coverage of a battery fire vs an oil fire.

The Tesla fire reminded me to try and get the latest safety info on batteries vs oil / gas fires as its not real clear to me yet and immediately this oil well fire pops up and surprised me. I hadn't heard one thing about it reminding me how common place those kind of fires are and uninteresting a week long one is outside of the local news.

I thought it would be interesting to post here to maybe spure some insightful discussion about the safety of each, instead I get reminded that bringing up such a thing is annoying to some of you and then get compared to Russian propaganda, but I'm the What-about-ist...

But you're not answering the question.

Why oil fires? At a minimum, I'd think that natural gas fires are at least somewhat related (as batteries are expected to form a similar role as _natural gas_ peaker plants).

I can't think of any relevant link between oil and electrical batteries with regards to this story about the Australian battery pack. I assume there's an underlying reason why you're comparing the two?

-------

Would it be relevant if I were to discuss the Fukushima Nuclear Disaster as part of this Australian news story? I think we both agree that Fukushima is a utility-scale environmental disaster related to energy grids. But even then, I wouldn't know exactly what the point would be to bring up Fukushima over this.

Just trying to give you an idea from where I'm coming from. Yeah, environmental disasters happen and they continue to happen constantly. But I'm trying to figure out why that's relevant to Australia and/or its battery pack.

---------

Are you implying that you want us to treat electrical battery installations as dangerously as oil fields and oil wells?

Where I live in FL,US one third of the peaker plants are oil from my understanding. There are plenty of oil and natural gas storage all over the US including at the peaker plants.

I am actually more interested in automobile fires, where both batteries and oil (refined in gasoline or diesel or even plastic) are the main worries, but I simply searched oil fire safety as I was focused more on power plants from this story.

Nuclear seems to have no problem getting reported on the news, so I didn't feel the need to bring it up although there is a lot to discuss there on real safety vs percieved.

Oh please. You're a Tesla fanatic and apologist that can't help but spam post in every single topic even remotely associated with electric cars, and the only thing you're doing here is trying to deflect from obviously bad press about your favorite company in the whole entire world.

The worrying part is assuming we're all on board with the implication that whataboutery is a bad or invalid form of discussion. So many times I see relevant lines of discussion pointlessly shutdown because someone shouts whataboutism as if it discredits the argument or point.

I wish we had better fire protection against these batteries.

As we're ramping up production and large packs become ubiquitous, in cars, factories, houses, we're really ill-equipped to combat their nasty tendency to burst into flames that are basically impossible to extinguish.

These things will burn if they fall below their minimal voltage, they will burn if you over-charge them, they will burn if you overheat them, they will burn if you constrain them too much, they will burn if you stab or mechanically stress them....

No fire services in the world appear to be equipped to put out those fires. At this stage, I'm wouldn't be surprised of some large-scale catastrophe brought by a defective cell in a larger pack somewhere.

> No fire services in the world appear to be equipped to put out those fires.

Huh? Fire services have already adapted to electric car fires - they got mobile rolling container tanks in which burning car carcasses are submerged: https://efahrer.chip.de/news/warum-die-feuerwehr-diesen-bmw-...

For battery fires in buildings, the situation is a bit different. Small scale (laptop, toys, powerbank, phone) fires can be extinguished with sand or common special metal fire fighting salts which melt and thus enclose the burning battery, depriving it of oxygen. While we're at it, it is always a good idea to keep a bucket of sand in your home - in case something battery-backed burns, empty the sand bucket over it.

Large scale aka battery pack fires are uncharted territory at the moment as these are a relatively new invention... but in general, it is imaginable that regulations will be adapted, to require putting a wall power bank in a separate room, similar to how natural-gas or diesel tanks are supposed to be installed.

> thus enclose the burning battery, depriving it of oxygen

Isn't one of the big problems with lithium battery fires that they produce their own oxygen?

From water, yes. If they're surrounded by something impermeable that isn't water, then that solves it.

> Isn't one of the big problems with lithium battery fires that they produce their own oxygen?

The problem with lithium fires is that if you douse them with water, the heat of the reaction (and/or lithium acting as a catalyst, my chemistry classes are a decade ago and my memory is hazy at best) splits the water into oxygen and hydrogen. The hydrogen is the real issue as it immediately combusts explosively.

According to testing they did on the previous generation of PowerPacks, it's extremely difficult for it to catch fire, and the damage tends to be contained within a single cell. It requires a substantial event like a high-energy puncture or external fire to actually set it alight. Very curious to see the investigation on this one.

> I wish we had better fire protection against these batteries.

Even better, let's work out chemistries that don't have the same level of risk. We already have LFP, which is an order of magnitude less likely to combust. IMO we really shouldn't be using traditional lithium ion batteries for things like grid storage or home batteries where weight isn't a prime consideration. I wouldn't want one in my house, I'd use LFP instead. Or even better, LTO, though it's not as cost effective. If I have to sleep above it...

If LFP is so good, why isn't it already used? Supply?

It is popular, but it does not have the most well-known battery company in the world standing behind it. Volume EVs in China use LFP, which has the effect of making those cells common and then available cheaply in the rest of the world. So DIYers love them, and there are lots of commercial products that rely on LFP.

Tesla doesn't have any incentive at this point to use LFP, they're selling every Powerwall they can make. The people who are buying them aren't price sensitive. The folks who are, buy LFP instead.

Elon Musk has recently stated that they are going to make a big shift to LFP batteries[1]. On the Tesla's quarterly earnings call last week, he stated that iron based batteries will likely eventually be used for almost all battery applications due to the low cost. Iron is cheap and sourced all over the globe. He said airplanes and ships will probably have to still use Nickel and/or Cobalt (but corrected himself later saying only airplanes will probably need them).

[1]https://www.torquenews.com/1/lfp-batteries-pros-and-cons-elo...

That is excellent news!

I was doing some research the other day on state-of-the-art and noticed LFP was gaining some ground on power density, getting some love from Tesla is a big win. Future looks bright.

The patent portfolio for LFP could be used for free only in China since about 2003. The patents expire everywhere next year.

We learned how to very safely handle big tanks of extremely flammable fluids with explosive volatile vapors. We also learned how to deal with a system that pipes extremely explosive flammable gas into millions of homes, and outlets with high voltage electricity everywhere. We'll learn how to handle these too.

Nothing capable of powering modern society or transport is going to be made of unicorn glitter. Stored energy is stored energy and there are always dangers with it.

I am curious if Tesla's batteries have had more fire issues or if they're just reported more because it's Tesla and that makes it click bait.

I think a bit of both. Tesla is right at the edge of the current capabilities in terms of storage density so it is kind of logical that they would be the ones to find out the hard way what the real physical limits are with respect to handling and safety, and they are also the system that is the most widely deployed. On top of that anything Tesla is automatically newsworthy.

The most practical suggestion about how to deal with the problem that I've seen pass by is to ensure a certain minimum spacing between packs to create 'fire breaks' beyond which one pack can not ignite the next one over, but this would waste rather a large amount of space and probably would not work in situations like parking lots and such.

Yeah, I believe the best handling procedures for a big tank of extremely flammable fluids with explosive volatile vapours that is sitting in a fire[0] is to run away - a long way away. Sure, put out the fire and cool down the tank too, if you can do that from a long way away.

[0] https://www.firerescue1.com/firefighter-training/articles/wh...

Even with good safety procedures and legislation, we still get the storage of flammable liquids catastrophically wrong occasionally: http://www.earthquakes.bgs.ac.uk/research/events/buncefield_...

LiIon batteries may burn occasionally, but at least for grid storage the effects are localized and tend not to be explosive.

Fire services are learning how to deal with batteries [1] and most fire departments have had to adapt to new fire hazards as society changes over time with automobiles, high-rise buildings etc. For grid storage however it doesn't seem unreasonable to require each site to have the ability to move a burning pack somewhere where it can be extinguished.

[1] https://cfpa-e.eu/container-puts-out-inextinguishable-fires-...

Moving a burning pack seems overkill and unnecessarily hazardous. Rather than moving the fire to containment, it'd be a lot safer and easier to move containment to the fire. All you'd need is a few fireproof walls stored on site which can be moved into place when needed.

I'm pretty sure the lessons learned from this incident will be incorporated into future builds (and conceivably even some retrofits to this Victoria build.)

Since the stock fire-fighting response to this is (according to the article) "Keep other battery packs close to the fire cool and let it burn itself out," (Lithium fires, eh!) I'm guessing that something like automatic (fire-triggered) cooling systems (if that's not already in place - the article made no mention of existing fire-response systems and mitigations) and possibly better access between the packs for fire-fighting machinery.

I'm pretty sure that the occasional fire is just part of the risk in building these extremely large agglomerations of Lithium cells, just as it is part of the risk in existing electrical substations. All things considered, I'm reasonably impressed that the fire remained as contained as it has.

> I'm reasonably impressed that the fire remained as contained as it has.

Indeed—though one must assume that this kind of catastrophic event was specifically predicted and planned for. So the fact that the fire didn't spread is really just evidence that they knew what they were doing.

I think we need to stop giving Tesla the benefit of doubt

Electric vehicles catch fire at a much lower rate than internal combustion engines.

It turns out what you said is also true of tubes full of gasoline.

Energy storage is hard. We'll need to iterate firefighting techniques to handle large batteries, just like we need to work on third-party repair for BEV and battery recycling programs. Fortunately these are all relatively straightforward engineering problems... small steps towards the sustainable future we must invent to survive as a species.

Intrinsically safe energy storage is an oxymoron. The objective is storing large amounts of energy to be released slowly over time - if there exists any possibility whatsoever of imperfect containment, there is (by definition) some probability that it will fail and release that energy all at once.

Can they be made to fail safely? Sure.

Can they be made to never fail? Not for what anyone's willing to pay for them.

Lithium batteries store about 0.5 MJ/kg. Compared to the specific energy of burning lithium metal (40 MJ/kg) the energy from the electricity producing redox reaction is pretty low.

In the case of lithium batteries. Our problem isn't the stored electric energy, it's the medium being a highly flammable alkaline metal. If we could somehow store that 0.5 MJ/kg in a non-flammable medium with similar energy density it would have 5% the specific energy of a similarly massing piece of hardwood.

Oh damn, TIL. You'd still have the issue of the stored energy in a non-flammable medium, but that 80x energy multiplier makes the problem distinct from any other energy storage I can think of. That's like keeping your diesel in a tank made entirely out of dynamite.

A candle stick has 100x the energy of a stick of TNT. I know which one I would rather be driving a truck full of.

Energy density is not the main danger factor.

> We'll need to iterate firefighting techniques to handle large batteries

Are there any firefighting techniques in use for handling car-size batteries?

Absolutely!

The state of the art is to have a container, fill it with water, and put the car in. You then wait a week or two.

For installations like this one which just burned down the answer is mostly: Don't let it catch fire! And if it does catch fire, make sure that you built it with large enough gaps between each unit that the fire can't spread, because you're not putting it out.

> The state of the art is to have a container, fill it with water, and put the car in. You then wait a week or two.

As a firefighter I can confirm this.

What worries me though is, that this might work well for cars, but what about batteries in homes, like Tesla powerwall and alike?

We do not yet have any plans for those, you can't "easily" attach them to a crane and lift them into a container. They are built in basements, garages and more. This will become challenging in the future...

Also, maybe "dont use tricky NMC batteries for stationary storage, when LFT can be cheaper and safer".

Still, hypothetically, can you use a (large) hose to fill a burning MegaPack container with water to cool it down? Maybe possible shorts in the other pods will make things worse?

LFP*?

These are also all negative externalities at this point.

Even straightforward engineering problems need funding to get engineers assigned to them.

To me it sounds pretty good right, the fire is controlled in three days no other batteries have caught fire because of this fire. hopefully they find out what is the cause and make changes.

> Something something federal gov want it's own probe before pushing forward

Why they didn't push before all that was even started ?? Is it now a rule in US that gov is told what could be potentially expected and gov "pushes forward" ? Like with military ships with only touch displays controls, planes with so big engines that body was not even imagined to have them attached, cable companies telling gov what coverage they have (but really not), etc, etc ?

In not so old day it was known that military guys are dumb and without imagination eg. discarding idea of Gutling's gun (until he started to gather attention somewhere else), or do not being interested in some fishy stealh jet technology... But now looks like civilian part of gov is completly and utterly dumb and clueless. And looks it's a word-wide phenomena...

EDIT: Forget to mention, that being grounded in reality is a + for military guys and that looks like thay found processes for some innovations, eg. DARPA. On the other hand looks civilians-in-power are intentionaly cherish chaos, eg. Congress, Senate, Parliament.

This is a total tangent from the article, and I understand what it actually entails, but I just love that there is a real office in the Australian government whose title is 'Shadow Energy Minister'. Sounds like something out of warhammer.

Lifepo4 for stationary storage makes so much more sense vs NMC.

I agree. To iterate, the main advantages I see are:

- Lower cost / kWh

- No need for Cobalt (and its controversial extraction)

- Better for the environment (from my basic understanding the materials are easier to obtain and less damaging)

- Longer cycle life (~5K vs 2K)

- No thermal runaway and less risk of catching fire

The main downside is that LFP batteries are heavier per kWh, but that should mostly be a non-issue for stationary storage.

The price of LFP batteries is dropping so fast. Consumers can buy a fully assembled LFP 12V 1300Wh battery for $399 on amazon [0]. Raw cells can be bought for 3-4x less from Chinese distributors.

[0] https://www.youtube.com/watch?v=74iBMHE3StM

It sounds like each of these big packs should have enough fire prevention systems to be able to put itself out in the case it does catch fire.

These packs have fire suppression built in, but it obviously failed.

The newer LiFePO4 packs don't require this fire suppression and should be a little bit cheaper, but not as dense.

Reply on news.ycombinator.com