Author Topic: Moss Landing Li battery fire.  (Read 2140 times)

0 Members and 2 Guests are viewing this topic.

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 9428
  • Country: fi
Re: Moss Landing Li battery fire.
« Reply #25 on: January 22, 2025, 07:09:51 pm »
They started out lithium-ion and very discretely changed over to LiFePO4.

LiFePO4 is lithium ion. What you say is like: "First we were served fruits, but then they were discretely changed to apples."

There is noting "discrete" here, different cathode chemistries and their trade-offs are all well-known.

Quote
I called them "lithium batteries" and there is still relevance for firefighters who shouldn't have to guess what kind of fire they are fighting, electricity included.

You confusing others by hallucinating about stuff you don't have any idea about (and not willing to learn about) is not relevant to anyone, least firefighters. Regardless of if the cathode chemistry is LFP, NCO or some other lithium ion chemistry - water is the way to cool them down.

Letting burn is always an option with any type of fire, of course.
 
The following users thanked this post: tom66

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7744
  • Country: ca
Re: Moss Landing Li battery fire.
« Reply #26 on: January 22, 2025, 08:21:03 pm »
What battery chemistry is in the Powerwall, Powerpack, Megapack, Megapack 2 etc.?
I think differences, the change is to less volatile iron phosphate lithium batteries is seen across the board.
As a firefighter arriving on scene, you say it's irrelevant- just pour on the water and cool it off? So why did the firefighters do nothing if it's that simple? Maybe you could work as a consultant to Vistra or the Monterey Fire Dept.

Paper is a bit interesting Toxic fluoride gas emissions from lithium-ion battery fires water increased the hydrogen fluoride production rate but not the total volume.

edit: Senate Bill No. 38 passed in 2023, about having an emergency action plan. Oops.
« Last Edit: January 22, 2025, 08:32:50 pm by floobydust »
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 10113
  • Country: gb
Re: Moss Landing Li battery fire.
« Reply #27 on: January 22, 2025, 08:33:14 pm »
What battery chemistry is in the Powerwall, Powerpack, Megapack, Megapack 2 etc.?
I think differences, the change is to less volatile iron phosphate lithium batteries is seen across the board.
As a firefighter arriving on scene, you say it's irrelevant- just pour on the water and cool it off? So why did the firefighters do nothing if it's that simple? Maybe you could work as a consultant to Vistra or the Monterey Fire Dept.

Paper is a bit interesting Toxic fluoride gas emissions from lithium-ion battery fires water increased the hydrogen fluoride production rate but not the total volume.
Fire fighting in a warehouse full of lithium cells is very different from firefighting in a warehouse full of lithium power packs. In one case they aren't chained into a voltage that could kill a firefighter with some conduction through the water they are using. In the other case they are.
 

Online neverendingstudent

  • Regular Contributor
  • *
  • Posts: 55
  • Country: us
Re: Moss Landing Li battery fire.
« Reply #28 on: January 22, 2025, 09:06:08 pm »
Another option I can think of (though I'm going to start by admitting there are a TON of complicating issues with this) would be at-battery compressed nitrogen storage in high volume.

Useless for li-ion batteries. They burn without external oxygen; they have their own internal oxidizer.

Really, li-ion fires and their prevention/mitigation is really well understood. All that is needed is investigating why the knowledge is not applied to practice (my guess: money), and discussing what should be done to enforce proper engineering (which is getting to politics).

Oh, I know about the issue of not needing oxygen, I was thinking specifically of the cooling aspect.  Water has more thermal capacity, but will start out at whatever room temperature is.  Compressed nitrogen, rapidly decompressed will produce much much lower outflow temperatures.  My perspective was to get as much as possible as cold as possible to arrest the catch-fire chain reaction of one cell to the next.  It would also not introduce the complicating factor of water, dirty with everything it is mixing with, being conductive and shorting out more batteries.
« Last Edit: January 22, 2025, 09:08:23 pm by neverendingstudent »
 
The following users thanked this post: Siwastaja

Offline floobydust

  • Super Contributor
  • ***
  • Posts: 7744
  • Country: ca
Re: Moss Landing Li battery fire.
« Reply #29 on: January 22, 2025, 09:11:38 pm »
What battery chemistry is in the Powerwall, Powerpack, Megapack, Megapack 2 etc.?
I think differences, the change is to less volatile iron phosphate lithium batteries is seen across the board.
As a firefighter arriving on scene, you say it's irrelevant- just pour on the water and cool it off? So why did the firefighters do nothing if it's that simple? Maybe you could work as a consultant to Vistra or the Monterey Fire Dept.

Paper is a bit interesting Toxic fluoride gas emissions from lithium-ion battery fires water increased the hydrogen fluoride production rate but not the total volume.
Fire fighting in a warehouse full of lithium cells is very different from firefighting in a warehouse full of lithium power packs. In one case they aren't chained into a voltage that could kill a firefighter with some conduction through the water they are using. In the other case they are.

It's saying containment is the only solution but I see nothing for that in these battery skids. They are just plopped down like Lego blocks, close to each other and lots of sheet metal that melts fast. No firewall or decent spacings.
Somewhere a myth exists on how these won't burn up. They are ending up like nuclear power plants. Where is some foam or dry chemical technology?
SpaceX has thermal tile tech, what it's too much extra cost or something.

The fire fallout so far is in the environmental consequences, for the farms and air/water pollution. It's quite a mess. Meanwhile Vistra stock... to the moon
 

Online nctnico

  • Super Contributor
  • ***
  • Posts: 28482
  • Country: nl
    • NCT Developments
Re: Moss Landing Li battery fire.
« Reply #30 on: January 22, 2025, 09:33:43 pm »
Another option I can think of (though I'm going to start by admitting there are a TON of complicating issues with this) would be at-battery compressed nitrogen storage in high volume.

Useless for li-ion batteries. They burn without external oxygen; they have their own internal oxidizer.

Really, li-ion fires and their prevention/mitigation is really well understood. All that is needed is investigating why the knowledge is not applied to practice (my guess: money), and discussing what should be done to enforce proper engineering (which is getting to politics).
One way to deal with it in an easy way is to put the batteries in a structure which can be flooded with water so the batteries become submerged. Just like they do with electric cars; dump them in a large bin filled with water and let it sit for a couple of days. With the batteries charged or not, it doesn't matter. The energy will need to be dissipated anyway.
« Last Edit: January 22, 2025, 09:35:42 pm by nctnico »
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 
The following users thanked this post: Siwastaja

Online tom66

  • Super Contributor
  • ***
  • Posts: 7374
  • Country: gb
  • Professional HW / FPGA / Embedded Engr. & Hobbyist
Re: Moss Landing Li battery fire.
« Reply #31 on: January 22, 2025, 11:45:51 pm »
What battery chemistry is in the Powerwall, Powerpack, Megapack, Megapack 2 etc.?

Megapack 1 is NCA/NMC.
Megapack 2 is LFP.
I'm fairly sure Powerwall/Powerpack are all NMC or NCA depending on the generation.
 

Offline SiliconWizard

  • Super Contributor
  • ***
  • Posts: 15893
  • Country: fr
Re: Moss Landing Li battery fire.
« Reply #32 on: January 22, 2025, 11:52:03 pm »
LiFePO4, while this chemistry is safer under normal (or barely beyond normal) conditions than basic Li-ion (but with lower energy density, you can't have it all), in case of fire (which may be caused by something else than the batteries), the result is basically the same.
 

Online tom66

  • Super Contributor
  • ***
  • Posts: 7374
  • Country: gb
  • Professional HW / FPGA / Embedded Engr. & Hobbyist
Re: Moss Landing Li battery fire.
« Reply #33 on: January 23, 2025, 12:12:54 am »
LFP batteries do require a considerably higher temperature to enter thermal runaway: around 350C for a BYD LFP battery, for instance.  This does make them considerably safer, since it may be possible to keep temperatures below that level with an appropriate sprinkler system, even if one cell or bank does become, er... "angry". 

NMC is typically around 180-200C for thermal runaway and NCA can be as low as 150C.

 
The following users thanked this post: Siwastaja

Online nctnico

  • Super Contributor
  • ***
  • Posts: 28482
  • Country: nl
    • NCT Developments
Re: Moss Landing Li battery fire.
« Reply #34 on: January 23, 2025, 12:23:56 am »
LFP batteries do require a considerably higher temperature to enter thermal runaway: around 350C for a BYD LFP battery, for instance.  This does make them considerably safer, since it may be possible to keep temperatures below that level with an appropriate sprinkler system, even if one cell or bank does become, er... "angry". 
I doubt that as the cells would need to be submerged in water and/or sufficient water ingress to get cooling where it is needed must be guaranteed by design.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 9428
  • Country: fi
Re: Moss Landing Li battery fire.
« Reply #35 on: January 23, 2025, 06:49:46 am »
What battery chemistry is in the Powerwall, Powerpack, Megapack, Megapack 2 etc.?

It doesn't matter. None of them contain lithium in a form that reacts with water, which was your original implication even if you now try to hide it as you realized how ridiculous it was.

Quote
I think differences, the change is to less volatile iron phosphate lithium batteries is seen across the board.

It is true that LFP batteries have higher thermal runaway onset temperature (so smaller risk of failure in small events; but it's still susceptible to chain event) and smaller release of energy during the event. So it's safer (than e.g. NCA or NMC) - but not even by a full order of magnitude. Safety (assuming it's not ruined by slacking on other safety measures "because it's safer") is one argument for using LFP. But really, they are used for other reasons; like cost of raw materials, longer cycle life at large DoD%... The compromise on energy density is not a big problem on stationary packs.

Quote
As a firefighter arriving on scene, you say it's irrelevant- just pour on the water and cool it off?

No, I say firefighters do not start googling about the battery chemistry and guessing that maybe it has metallic lithium which reacts with water because floobydust speculates so on the internets.

At least hopefully.

What they need is, like with any large industrial setting, a safety plan, a firefighting plan, which takes into account all the details, like: how to ensure no dangerous voltages are present? Where to spray the water (if at all - maybe it's better to let it burn)? What is the risk level of gases emitted? Risk of firefighters being engulfed if the fire suddenly spreads - how fast can it spread? Water being used to cool the neighbor units to prevent cascading effect, and if so, how? Is there some specific "spray here for emergency cooling" hole? Or just cool the outsides of the containers?

Quote
Maybe you could work as a consultant to Vistra or the Monterey Fire Dept.

People more involved in lithium ion chemistry than me most definitely work with fire depts. They don't google for speculation on internet forums.
« Last Edit: January 23, 2025, 07:26:20 am by Siwastaja »
 

Online Marco

  • Super Contributor
  • ***
  • Posts: 7080
  • Country: nl
Re: Moss Landing Li battery fire.
« Reply #36 on: January 23, 2025, 08:17:09 am »
One way to deal with it in an easy way is to put the batteries in a structure which can be flooded with water so the batteries become submerged. Just like they do with electric cars; dump them in a large bin filled with water and let it sit for a couple of days. With the batteries charged or not, it doesn't matter. The energy will need to be dissipated anyway.

A bit of distance helps more. The most practical design is to keep the walls of neighbouring modules cool and let it burn.
 
The following users thanked this post: Siwastaja

Online tom66

  • Super Contributor
  • ***
  • Posts: 7374
  • Country: gb
  • Professional HW / FPGA / Embedded Engr. & Hobbyist
Re: Moss Landing Li battery fire.
« Reply #37 on: January 23, 2025, 09:30:56 am »
LFP batteries do require a considerably higher temperature to enter thermal runaway: around 350C for a BYD LFP battery, for instance.  This does make them considerably safer, since it may be possible to keep temperatures below that level with an appropriate sprinkler system, even if one cell or bank does become, er... "angry". 
I doubt that as the cells would need to be submerged in water and/or sufficient water ingress to get cooling where it is needed must be guaranteed by design.

Well, we don't know if the building had a sprinkler system and if it did how effective it was intended to be, but I would have thought water deluge in the area around the affected cell bank would prevent adjacent cell banks from getting up to the critical temperature of 350C.  As others say, all about partitioning between adjacent cells. Rule of thumb is 1kWh NMC burns with roughly 10-15kWh chemical energy (unsure on LFP but likely similar), so your cooling system needs to be able to remove enough of that heat to stop adjacent banks from combusting. Even if you lose a whole row of modules, it's better than the building!
 

Online Marco

  • Super Contributor
  • ***
  • Posts: 7080
  • Country: nl
 

Online tom66

  • Super Contributor
  • ***
  • Posts: 7374
  • Country: gb
  • Professional HW / FPGA / Embedded Engr. & Hobbyist
Re: Moss Landing Li battery fire.
« Reply #39 on: January 23, 2025, 10:31:35 am »
1kWh NMC burns with roughly 10-15kWh chemical energy
Really?

https://www.fire.tc.faa.gov/pdf/TC-TN16-22.pdf
https://iopscience.iop.org/article/10.1149/1945-7111/ac3c27/pdf

If I'm skimming those papers correctly, they're saying around 1-1.5x stored energy is released as heat, so considerably lower?  That does seem a bit too low to be believed, so we're probably talking about different things.

The figure I cite is from an electric car fire study, where the battery is assumed to be completely consumed by fire in the ensuing thermal runaway and subsequent vehicle fire.  In other words: how much energy did that battery contribute to the fire? For a typical EV battery it turned out to be equivalent to 70-100 litres of petrol.  (60kWh EV battery = 216MJ stored energy, ~2.1-3.2GJ combustion energy). I'd assume that figure does include things like module interconnects, PCBs, insulators, plastics, casing etc. most of which would be shared with a PowerPack type system. It came about in the EV fires thread we had a few months ago, I'll have to dig through it later, but the bottom line was that the total contribution was around 10x more than the stored energy.
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 9428
  • Country: fi
Re: Moss Landing Li battery fire.
« Reply #40 on: January 23, 2025, 11:07:50 am »
The number I remember is roughly 2x. Could be wrong.

Don't underestimate the contribution of other combustible materials. I remember a calculation where it was shown that for shipping li-ion cells, cardboard (+ plastic, as in foams) packaging material is a significant contributor to the total energy release in case of fire.

Thinking that li-ion cell is some super concentrated energy storage is exaggerated. For example, car fires where just plastic parts in engine compartment, tires, brake fluids etc. burn look quite violent, even if no gasoline or battery pack is involved at all. Li-ion fires (e.g. in aircraft cargo) are dangerous because they don't extinguish by simple removal of air (replacement with vacuum or nitrogen/argon gas, or CO2 or powder extinguisher). And they set other stuff in fire. Not because the energy release is exceptionally high.
« Last Edit: January 23, 2025, 11:10:16 am by Siwastaja »
 

Online tom66

  • Super Contributor
  • ***
  • Posts: 7374
  • Country: gb
  • Professional HW / FPGA / Embedded Engr. & Hobbyist
Re: Moss Landing Li battery fire.
« Reply #41 on: January 23, 2025, 11:33:23 am »
Right, and that was the interesting thing from that study, which was the interior and plastics in the vehicle contributed 90% of the energy, so it didn't really matter if you had a 50 litre fuel tank or a 60kWh battery, the net result was the same. Yes, the battery is a bastard to extinguish once it is aflame, but once the interior components are involved, it doesn't really matter from a thermal loading perspective.  I imagine much the same is true for a battery storage facility, there will be a lot of non-battery components in there that will become involved.
 

Online Marco

  • Super Contributor
  • ***
  • Posts: 7080
  • Country: nl
Re: Moss Landing Li battery fire.
« Reply #42 on: January 23, 2025, 12:44:36 pm »
If I'm skimming those papers correctly, they're saying around 1-1.5x stored energy is released as heat, so considerably lower?  That does seem a bit too low to be believed, so we're probably talking about different things.

I'll just believe the well controlled calorimeter studies for the moment. Calorimetry is hard and it gets far harder when things get larger.
 

Offline Simmed

  • Regular Contributor
  • *
  • Posts: 87
  • Country: 00
Re: Moss Landing Li battery fire.
« Reply #43 on: January 23, 2025, 01:41:38 pm »
this paper is too large for attachment 11mb

https://www.mdpi.com/2313-0105/11/1/24

the chart data shows the 20s "explosion" window for NCM523, which gives a rough estimate of about 390k joules in 20s
« Last Edit: January 23, 2025, 02:04:49 pm by Simmed »
V=I.R  Q=h.A.(dT) q=(dT).p  Q=C.V  F=m.a  F=q.v.B.(sinθ)
 

Online tom66

  • Super Contributor
  • ***
  • Posts: 7374
  • Country: gb
  • Professional HW / FPGA / Embedded Engr. & Hobbyist
Re: Moss Landing Li battery fire.
« Reply #44 on: January 23, 2025, 02:26:00 pm »
I was speaking to a friend the other day who works in automotive and he mentioned an interesting new test for EVs.  EV batteries are subjected to direct fire exposure for 2 minutes.  The expectation is that the battery does not enter thermal runaway.  The vehicle is allowed to use the cooling system to protect the battery, including localised cooling to protect specific areas of the pack. UN ECE R100.  I wonder if such a regulation will come into play for stationary storage.
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 10113
  • Country: gb
Re: Moss Landing Li battery fire.
« Reply #45 on: January 23, 2025, 02:37:47 pm »
I was speaking to a friend the other day who works in automotive and he mentioned an interesting new test for EVs.  EV batteries are subjected to direct fire exposure for 2 minutes.  The expectation is that the battery does not enter thermal runaway.  The vehicle is allowed to use the cooling system to protect the battery, including localised cooling to protect specific areas of the pack. UN ECE R100.  I wonder if such a regulation will come into play for stationary storage.
That sounds like a very strange test. Testing batteries for their behaviour in a fire sounds good, but why only 2 minutes? When is a real EV battery ever going to be exposed to a fire for only 2 minutes?
 

Online tom66

  • Super Contributor
  • ***
  • Posts: 7374
  • Country: gb
  • Professional HW / FPGA / Embedded Engr. & Hobbyist
Re: Moss Landing Li battery fire.
« Reply #46 on: January 23, 2025, 02:54:05 pm »
I was speaking to a friend the other day who works in automotive and he mentioned an interesting new test for EVs.  EV batteries are subjected to direct fire exposure for 2 minutes.  The expectation is that the battery does not enter thermal runaway.  The vehicle is allowed to use the cooling system to protect the battery, including localised cooling to protect specific areas of the pack. UN ECE R100.  I wonder if such a regulation will come into play for stationary storage.
That sounds like a very strange test. Testing batteries for their behaviour in a fire sounds good, but why only 2 minutes? When is a real EV battery ever going to be exposed to a fire for only 2 minutes?

My guess would be in case of a fuel spill on a roadway.  You need to give the people in the car time to evacuate in a sudden situation.  I think they assume that continuous exposure to fire will cause the pack to enter runaway so there's no way to protect against this.  It might also be trying to address the rate of fire spread in parking garages etc.

For a point of comparison, petrol tanks only have to withstand 2 minutes of flame before catastrophic failure, so it's a comparable objective.

This document is also interesting: https://wiki.unece.org/download/attachments/25270993/EVSTF-03-29e.pdf?api=v2

Seems that the quickest time to thermal runaway was around 150s for the small PHEV, but most vehicles took 300-400s. 
 
The following users thanked this post: Siwastaja

Online nctnico

  • Super Contributor
  • ***
  • Posts: 28482
  • Country: nl
    • NCT Developments
Re: Moss Landing Li battery fire.
« Reply #47 on: January 23, 2025, 03:58:18 pm »
I was speaking to a friend the other day who works in automotive and he mentioned an interesting new test for EVs.  EV batteries are subjected to direct fire exposure for 2 minutes.  The expectation is that the battery does not enter thermal runaway.  The vehicle is allowed to use the cooling system to protect the battery, including localised cooling to protect specific areas of the pack. UN ECE R100.  I wonder if such a regulation will come into play for stationary storage.
That sounds like a very strange test. Testing batteries for their behaviour in a fire sounds good, but why only 2 minutes?
Probably to give passengers (extra) time to get out of the car. Safety regulations are only there to protect human life.
There are small lies, big lies and then there is what is on the screen of your oscilloscope.
 

Offline coppice

  • Super Contributor
  • ***
  • Posts: 10113
  • Country: gb
Re: Moss Landing Li battery fire.
« Reply #48 on: January 23, 2025, 05:14:24 pm »
I was speaking to a friend the other day who works in automotive and he mentioned an interesting new test for EVs.  EV batteries are subjected to direct fire exposure for 2 minutes.  The expectation is that the battery does not enter thermal runaway.  The vehicle is allowed to use the cooling system to protect the battery, including localised cooling to protect specific areas of the pack. UN ECE R100.  I wonder if such a regulation will come into play for stationary storage.
That sounds like a very strange test. Testing batteries for their behaviour in a fire sounds good, but why only 2 minutes?
Probably to give passengers (extra) time to get out of the car. Safety regulations are only there to protect human life.
In a crash bad enough for start a serious fire the passengers aren't typically in the greatest shape for getting out fast.
 

Online tom66

  • Super Contributor
  • ***
  • Posts: 7374
  • Country: gb
  • Professional HW / FPGA / Embedded Engr. & Hobbyist
Re: Moss Landing Li battery fire.
« Reply #49 on: January 23, 2025, 05:35:31 pm »
In a crash bad enough for start a serious fire the passengers aren't typically in the greatest shape for getting out fast.

This type of test is specifically aimed at external fire sources.  I imagine for instance a multi-car pileup where fuel is spilled on/under your vehicle.

Vehicle crash resulting in battery fire is specifically tested too, essentially, it can't happen in any of the standard tests where passenger survival would be expected.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf