Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

How Tesla's battery design avoided battery fire from single defective cell

stopcrazypp

Well-Known Member
Dec 8, 2007
10,977
6,088
This is fairly relevant given the huge LG Chem recalls, but there was an article with excerpt from the recent book about Tesla's history, that describes how Tesla designed the pack in the Roadster to avoid one defective cell leading to the entire pack catching on fire.

Key point:
Eventually, Straubel began to narrow in on a solution. If they couldn’t keep a cell from warming, maybe they could keep it from reaching the point where it set off a chain reaction. Through trial and error, the team realized that if they had each cell lined up a few millimeters from its neighbor, snaked a tube of liquid between them, and dumped a brownie-batter-like mixture of minerals into the resulting battery pack, they could create a system that contained overheating. If a defective cell within began to overheat, its energy would dissipate to its neighboring cells, with no individual cell ever reaching combustibility.
Hitting the Books: How Tesla engineers solved the problem of exploding EV batteries | Engadget

I would take with a grain of salt what was quoted as said or the people involved (given some accounts from the book have been disputed, for example Tim Cook ever talking with Elon Musk), but the general idea matches the pack design Tesla ended up with.

Saw this linked in some discussion about Bolt's pack design (which extends to other pack designs using similar cells). Basically with the large cell designs, while the probability of a single cell catching fire may be lower (due to lower amount of cells), due to the size of the cells, being in a pouch with no solid metal housing per cell, and how they are packed together, a single defective cell pretty much guarantees at least an entire module catches on fire, if not the whole pack.

Other factors that may come into play with LG's case specifically is the higher energy density and chemistry type (NMC). The Bolt had 60kWh with 288 cells or 208 Wh/cell, and the Kona Electric had 64kWh with 294 cells or 218 Wh/cell. The Volt had a similar cell design, but used LMO and only 16kWh from 288 cells, or 56 Wh/cell. The Leaf had 24 kWh LMO and 192 cells or 125 Wh/cell

The other factor may be LG's patented stack and fold design. This may introduce defects that happen only after the cell layers have been put into the pouch and would not be easy to catch with QC (would probably need X-raying each cell). Leaf used a more conventional stack design which probably is less susceptible to this kind of defect.
 
  • Like
Reactions: GSP

Earl

Member
Jan 22, 2014
251
308
USA
Key point:
This is pretty much how Eberhart described Tesla's batteries at Tesla's initial rollout in Santa Monica. I think he used words more like 'flame suppressant goo" instead of "brownie batter" but the idea was the same and it was plausible enough (along with other due diligence) that we bought a car a few weeks later.
Eberhart (founding president of Tesla) was terrified of fires. The initial Roadster wall charging stations had smoke detectors that automatically shut them off.
 

Rusty1

Member
Sep 13, 2017
141
139
Ooltewah, TN
I think this battery design is hands down the best design. Significantly better than the LG design. GM rushed to beat the Model 3 to market. Now look at their recall and reputation.

All EV supporters should take this seriously. Burning peoples houses down is a big impediment to ev adoption. I would argue greater than range anxiety.
 
  • Helpful
Reactions: Brando

mongo

Well-Known Member
May 3, 2017
13,473
42,104
Michigan
Then why this fire if Tesla can suppress a single cell failure?


Also, there seems to be a higher likely hood of fire in the 85kw batteries. Remember Tesla limiting their supercharging. I only know of one Model 3/Y fire, but have heard of multiple S fires.
There is no data presented in that article to indicate the pack was at fault. Given the charging interrupted notification, I would be looking at the house/ wall connector AC wiring or vehicle's charger.
Also, regarding the article: the large explosions would be the tires or airbags. There are numerous cases of gas cars catching fire when parked.

There have been other cases where the Tesla was a fire victim, not the cause:
 

Earl

Member
Jan 22, 2014
251
308
USA
Burning peoples houses down is a big impediment to ev adoption.
Cities used to not allow garages to be attached to houses because gasoline cars were so dangerous. It took a while before the industry learned how to avoid gasoline fires to the point where society accepted them as well.
It's all part of the innovation process.
 
  • Helpful
  • Like
Reactions: Brando and Rusty1

stopcrazypp

Well-Known Member
Dec 8, 2007
10,977
6,088
Then why this fire if Tesla can suppress a single cell failure?


Also, there seems to be a higher likely hood of fire in the 85kw batteries. Remember Tesla limiting their supercharging. I only know of one Model 3/Y fire, but have heard of multiple S fires.
No evidence given at all that suggests the fire was from single cell failure. That whole article is behind a paywall, but the snippet available cites the electrical system as a possible cause. The car was charging at the time. We've seen wall connections catch fire in the past, so that is no unusual.

So far the only proven cases have been ones where there was an impact of some sort. The S was more vulnerable to this due to it being designed for battery swapping (so the battery pack is more exposed, they even had to add more armor to it). The Model 3/Y threw that design out the window (no longer designed for swapping) so it is less vulnerable, but there have still been cases of impact causing fire (like the Shanghai case where the car hit a manhole over and pack caught on fire).

However, we know for certain from the Bolt fires (and similar Kona fires) are from cell defects (most likely single cell, with only 288 cells the probability of two on the same car and so many cases is highly unlikely). Multiple cases of the car finished charging or not even charging at all or plugged in. Some of the cases the battery was intact enough to determine fire started in the battery. Defective cells have been found in the factories that passed QC.
Great summary here of the cases:
Everything we know about the Chevy Bolt EV fires
 
Last edited:
  • Like
Reactions: GSP

Rusty1

Member
Sep 13, 2017
141
139
Ooltewah, TN
I would like to know more about the Shanghai fire causes. Tesla changed supercharging limitations based on this fire. There must have been something causing concern in the 85 packs.
 

mongo

Well-Known Member
May 3, 2017
13,473
42,104
Michigan
I would like to know more about the Shanghai fire causes. Tesla changed supercharging limitations based on this fire. There must have been something causing concern in the 85 packs.
That was an S which doesn't have the goo between the cells in its modules. Tesla identifies reason behind Model S fire in Shanghai, issues update to fix issue
Here is an independent dive into BMS software changes by @wk057 :
Explaining Changes post-firmware 2019.16 Regarding Range Loss | wk057's SkieNET
 
  • Like
Reactions: GSP and Rusty1

Brando

Active Member
Sep 27, 2016
3,006
2,244
Bainbridge Island, WA
Vehicle fires are common, generally. According to the National Fire Protection Association, there were 212,500 vehicle fires that caused 560 civilian deaths, 1,500 civilian injuries and $1.9 billion in direct property damage in the U.S. in 2018.

seems Tesla is less dangerous with batteries
 

mongo

Well-Known Member
May 3, 2017
13,473
42,104
Michigan
Thanks for the info. Which battery packs didn’t have the goo between the cells?
Non goo packs are in the original Roadster and pre-Plaid S&X (plaid/ new LR probably do, but I have not seen a tear down yet)
3/Y with the four module setup do have goo.
 

Rusty1

Member
Sep 13, 2017
141
139
Ooltewah, TN
Non goo packs are in the original Roadster and pre-Plaid S&X (plaid/ new LR probably do, but I have not seen a tear down yet)
3/Y with the four module setup do have goo.
Great info. Thanks. It must be working. Very few Model 3/Y fires.
 
Last edited:

mongo

Well-Known Member
May 3, 2017
13,473
42,104
Michigan
Now - how is Tesla preventing fires in the LFP pouch battery packs they are importing from China?
Has pouch been confirmed? LG was adding 2170 capacity a while back.
Assuming it is a pouch design, the use of more smaller pouches with rigid thermal containment vessels provides fault isolation and cooling.
Aluminum thermal channel, pouch, thermal channel, pouch ... thermal channel. The thermal channels could act as ribs for the pack 9with added complexity).
In general, pouch stack are used in multiple applications. but it's a lot easier if the power draw is lower (less thermal issues). The stacks are placed under compression to prevent expansion, sort of like prestressed concrete.
 

stopcrazypp

Well-Known Member
Dec 8, 2007
10,977
6,088
  • Informative
Reactions: GSP and mongo

stopcrazypp

Well-Known Member
Dec 8, 2007
10,977
6,088
I would like to know more about the Shanghai fire causes. Tesla changed supercharging limitations based on this fire. There must have been something causing concern in the 85 packs.
Here's the link to the Tesla statement on that fire, although unfortunately it's not in text format and can't be easily translated:
微博
However from the Teslarati report linked previously, the fire seemed to have been isolated to a single module and didn't spread to the the other modules (at minimum it didn't burn the entire pack down or I don't see how they would have been able to make that determination). I tried to find aftermath pictures but didn't find any to verify though. From the earlier incidents of pack puncture Elon mentioned there are firewalls built into the Model S pack to prevent fire from spreading from one module to the next.
Model S Fire

The major difference with the 85kWh packs is there is a double stacked section in the front of the car that is in use (16 modules total), while the 60kWH packs don't have that part (14 modules). Given the debris tends to hit the front modules first, that may explain why the 85kWh packs are more vulnerable in general (at least before they added armor).

If the BMS issue (as updated by software) was the cause, that could have affected an entire string of batteries (all 72 cells in parallel would be affected) and it wouldn't just be an incident with one cell (something the original design was designed to prevent from escalating). With 72 cells affected, even if not all cells fail, having a decent bunch fail would be beyond the design intent, which was only to prevent a single internal cell short from propagating.
 
  • Like
Reactions: GSP and mongo

stopcrazypp

Well-Known Member
Dec 8, 2007
10,977
6,088
Non goo packs are in the original Roadster and pre-Plaid S&X (plaid/ new LR probably do, but I have not seen a tear down yet)
3/Y with the four module setup do have goo.
That is not quite accurate, although verifying this on the Roadster was extremely hard (given teardowns to cell level is extremely rare, esp. given the car itself was rare in the first place). The Roadster also had goo, but not on the full cell, only as high as the cooling ribbon (which was not the whole height of the cell apparently).
It is in 7:10 in this video, the blue "thermal potting compound":

The Model 3 instead covered the whole cell (the ribbon appeared to do so too):
Model 3 Pack: Intumescent goo?
 
  • Informative
Reactions: mongo

Products we're discussing on TMC...

About Us

Formed in 2006, Tesla Motors Club (TMC) was the first independent online Tesla community. Today it remains the largest and most dynamic community of Tesla enthusiasts. Learn more.

Do you value your experience at TMC? Consider becoming a Supporting Member of Tesla Motors Club. As a thank you for your contribution, you'll get nearly no ads in the Community and Groups sections. Additional perks are available depending on the level of contribution. Please visit the Account Upgrades page for more details.


SUPPORT TMC
Top