According to recent reports[1], Tesla is transitioning to using Iron Phosphate (LFP) Li-ion cells in it's Megapack stationary storage products. Unlike nickel-based cells, LFP is highly resistant to fire and thermal runaway.
Presumably, this particular installation was still using nickel-based cells.
It's been... interesting watching Tesla migrate from their more oddball technical decisions to using the same tech as the rest of the industry. LFP is a really conventional tech for stationary storage as well as vehicle applications with less demanding power-to-weight requirements. They also seem to have mostly switched from the AC motors that inspired their name to permanent magnet brushless motors like the rest of the car industry.
Conventional wisdom is that grid storage product was offered because battery production was outstripping what they could ship in cars at the time.
Remember that heavy industry isn't like software, you don't get to Just Decide to match the product to the market and switch when it makes sense. You make a bet years in advance on resource contracts and factory capacity, and if the needles don't match you need to find some way to generate revenue from what you actually have, not what you want to produce.
But yes: LiFePO4 has a high internal resistance (which is bad for applications with high current loads) and somewhat poor energy density (bad for mobile applications), but is significantly cheaper at scale due to the elimination of Cobalt and has an absolutely massive cell durability advantage (GREAT for grid storage).
But you won't see phosphate batteries in anything but budget/commuter EVs. They aren't something Tesla is going to bet on for their main production choices, but absolutely something you'd expect to see them optimize as they scale.
I'd agree that LFP is becoming conventional in stationary storage, but it's still relatively unusual in passenger vehicles outside of China. As far as I know, Tesla is unique amongst "western" automakers in using LFP cells in some of its cars.
Speculation, but I think this is likely to happen at some point. Perhaps when there is a new model of Powerwall introduced.
LFP makes sense in many ways for stationary storage, but it does have lower energy density, so you couldn't keep the same kWh in the same form factor without some changes.
>Who's paying for the transition? The customer or Tesla?
Since nearly all Tesla revenue comes from Tesla customers I would imagine the customers will be ultimately footing the bill.
>Was this comment controversial in any way?
Honestly it comes off as cynical and a little dumb -- which is why it might be grey. If a business agrees to a cost -- it will be the customers paying for that cost. Your peevish "Who is paying for this" tone suggests you're being indignant as though you're being asked to cough up some dollars here.
And like reddit asking why you're being downvoted will lead to more downvotes.
I think it's a fair question - hardware companies have recalled and replaced dangerous components before.
I got a free replacement battery from Apple in the mid 2000's when they said that a batch of their batteries had a "small fire risk"...
If I'm being cynical, I guess it depends if their line is:
"the old battery is dangerous which is why we've changed the materials we use"
or
"we're changing the materials we use, and that has this interesting side effect that totally coincidentally fixes a dangerous issue with out old batteries"
What does this question even mean? What does paying for mean in this context? Are you asking if Elon takes the cost out of his personal bank account? Or if they send every tesla customer a bill for new battery tech?
What is the difference between the customer or tesla paying?
I don't mind your question, but I find the answer is undefined.
It's not as if someone somewhere sits down and makes this choice. Tesla charges prices the market wants to pay, at volumes Tesla can manufacture, and tries to minimize its expenses. What happens in the end in terms of profit and loss is subject to literally thousands of variables, not just a change in a battery pack tech.
This is especially the case for an industry that's heavily into R&D and innovation.
You get downvoted because the question is redundant. It's like asking who is paying for the next gen iPhone or whatever. You buy the product as is and when a new better generation of the same product comes out you don't get a free upgrade because you bought the old version.
The fire spread to a second pack. You can see in the images that they are laid out in pairs. So two out of 80 burned completely, and we don't know what damage was caused to the others. One at least looks pretty singed.
The bad thing isn't the fire itself. It required multiple fire units to be on site at all times. That's resources you can't use anywhere else for that time.
They’re located in high voltage switchyards, because they basically have to be - and the nearest bushland is separated by a comfortable 20km of pasture. You can read up on Powercor and Augrid’s bushfire mitigation work if you’re having trouble sleeping.
Any more likely than a gas/diesel fire? I'd imagine most of the fire risk is at the start, when first responders haven't arrived yet. Once they arrive any wildfire risk is minimal because it can be immediately put out. In that respect there probably isn't a big difference between a gas fire that burns for 1hr and a battery fire that burns for 4 days.
It's interesting how closely placed the two center rows are. That seems to invite a spread of the fire. The land looks pretty remote/rural, so I can't imagine space is a constraint. I wonder if there's a technical reason for them to be so close together.
Bigger site costs more, requires more fencing, more concreting, more wiring, more security staff and cameras, etc.
I expect someone did the maths and decided that the low (and probably insured) risk of a fire taking out two rather than one container outweighed additional concrete/fencing/land/wiring.
It happens, unless there's something surprising/unusual to be found that caused it like remnants of oily rag, or empty bottle marked 'molotov cocktail', I assume more time & energy will be spent on 'are the existing racks/containers/etc. enough to contain and isolate fires' (or no they're not, why, we thought they were).
If a phone battery say catches fire it's generally not such a big deal - it's not surrounded by a million other phone batteries, and won't take four days to burn out.
Yeah, I thought when I opened HN without the downvote option I was so happy (compared to browsing reddit). But then I just realise that button will appear aft 500 karma lol.
These things don't look like shipping containers. They appear to be the same size/shape and have the proper attachment points to be handled as shipping containers, but are they actual containers?
Even Tesla wouldn't go so far as to mount batteries inside standard shipping containers. They might make ok housing but with all the thermal/electrical issues a bespoke design would be appropriate. The money saved by using a stock container is nothing compared to the cost of the product installed inside.
Who claimed that these were shipping containers? It comforms to the applicable ISO standards, it is transported like a shipping container. It is a shipping container
"They found a 13-tonne lithium battery inside a shipping container was fully involved and crews wearing breathing apparatus worked to contain the fire and stop it spreading to nearby batteries."
A fire literally inside a shipping container is a very specific thing that firefighters train for. There are even some unique tools for handling them (spray nozzles inserted through drilled holes.)
Readit News
Cached: https://web.archive.org/web/20210802083001/https://news.cfa....
Google Cache: https://webcache.googleusercontent.com/search?q=cache:J83v_Z...
https://news.ycombinator.com/item?id=28034815 (90 points/80 comments)
Presumably, this particular installation was still using nickel-based cells.
[1] https://cleantechnica.com/2021/05/11/tesla-transitions-to-lf...
Remember that heavy industry isn't like software, you don't get to Just Decide to match the product to the market and switch when it makes sense. You make a bet years in advance on resource contracts and factory capacity, and if the needles don't match you need to find some way to generate revenue from what you actually have, not what you want to produce.
But yes: LiFePO4 has a high internal resistance (which is bad for applications with high current loads) and somewhat poor energy density (bad for mobile applications), but is significantly cheaper at scale due to the elimination of Cobalt and has an absolutely massive cell durability advantage (GREAT for grid storage).
But you won't see phosphate batteries in anything but budget/commuter EVs. They aren't something Tesla is going to bet on for their main production choices, but absolutely something you'd expect to see them optimize as they scale.
LFP makes sense in many ways for stationary storage, but it does have lower energy density, so you couldn't keep the same kWh in the same form factor without some changes.
Funny this got downvoted. Was this comment controversial in any way?
Since nearly all Tesla revenue comes from Tesla customers I would imagine the customers will be ultimately footing the bill.
>Was this comment controversial in any way?
Honestly it comes off as cynical and a little dumb -- which is why it might be grey. If a business agrees to a cost -- it will be the customers paying for that cost. Your peevish "Who is paying for this" tone suggests you're being indignant as though you're being asked to cough up some dollars here.
And like reddit asking why you're being downvoted will lead to more downvotes.
I got a free replacement battery from Apple in the mid 2000's when they said that a batch of their batteries had a "small fire risk"...
If I'm being cynical, I guess it depends if their line is:
"the old battery is dangerous which is why we've changed the materials we use"
or
"we're changing the materials we use, and that has this interesting side effect that totally coincidentally fixes a dangerous issue with out old batteries"
What is the difference between the customer or tesla paying?
It's not as if someone somewhere sits down and makes this choice. Tesla charges prices the market wants to pay, at volumes Tesla can manufacture, and tries to minimize its expenses. What happens in the end in terms of profit and loss is subject to literally thousands of variables, not just a change in a battery pack tech.
This is especially the case for an industry that's heavily into R&D and innovation.
> a 13-tonne lithium battery inside a shipping container
The other ~99 container battery packs survived.
https://www.theverge.com/2021/7/30/22602411/neoen-tesla-mega...
I expect someone did the maths and decided that the low (and probably insured) risk of a fire taking out two rather than one container outweighed additional concrete/fencing/land/wiring.
actually I still can't find any explanation why the fire starts at all. Or maybe they haven't found out yet
Edit: I don't get the downvotes, I am from Melbie, nothing is controversial here
If a phone battery say catches fire it's generally not such a big deal - it's not surrounded by a million other phone batteries, and won't take four days to burn out.
Unless it's in your pocket!
So same same thing
These things don't look like shipping containers. They appear to be the same size/shape and have the proper attachment points to be handled as shipping containers, but are they actual containers?
Even Tesla wouldn't go so far as to mount batteries inside standard shipping containers. They might make ok housing but with all the thermal/electrical issues a bespoke design would be appropriate. The money saved by using a stock container is nothing compared to the cost of the product installed inside.
"They found a 13-tonne lithium battery inside a shipping container was fully involved and crews wearing breathing apparatus worked to contain the fire and stop it spreading to nearby batteries."
A fire literally inside a shipping container is a very specific thing that firefighters train for. There are even some unique tools for handling them (spray nozzles inserted through drilled holes.)