Another Tesla fire, this time while supercharging

[MrSlack]

Yep Potassium Oxide. Seen first hand what a mess that makes of stuff. Not pretty.

[aargee]

IIRC the Brits stopped building warships out of aluminium after the Faulkland conflict due to the tendancy
for it to burn (self sustained) given the right circumstances.

[Rick Law]

It may look rather bad, but it is not a total lost.  I distinctly see two used alloy wheels in good condition that could be cleaned and redeployed...

Back to the serious side.  Not having done an actual count, Tesla fire seem to be occurring at about one per 3 to 6 months.

I wonder how is the rate of total-destruction-by-fire per year per 1000 cars as compared to natural gas vehicle and as compare to gasoline vehicle's rate.

My gut feel is that because of it being new and expensive, the destruction of Telsa is fully reported worldwide, whereas gasoline cars burning off the far corners are likely ignored in the world stage.  So probability of fire is highly exaggerated.   Anyone know of hard-data out there?

[tom66]

Back to the serious side.  Not having done an actual count, Tesla fire seem to be occurring at about one per 3 to 6 months.

This doesn't follow.

AFAIK there have been seven Model S fires, but it's been released for just over three years now

1. road debris #1
2. mexico incident (very high speed collision through wall), fire occurred about 30 minutes afterwards
3. road debris #2
4. charging fire #1 (house wiring appeared to be cause - car was badly damaged but not source of fire)
5. fire after car split in half (thief stole car from LA service centre, crashed it 100+mph into traffic light pole, survived accident after flying out of car without a seatbelt, and died a week later in hospital)
6. fire after car went off the road and fell 500ft down (driver died - suspected suicide)
7. this incident (charging fire #2)

#1 and #3 were fixed by Tesla applying shield to car
#4 was fixed by software update adding arcfault and volt drop detection
#2, #5 and #6 are reasonable occurrences given the circumstances
#7 is the unknown as it appears spontaneous and while supercharging; IMO this is the most concerning one
 
7 fires / 117,000 cars is pretty low, especially when you consider that most appear to have been from "blunt-force" trauma to the battery pack. Only two spontaneous fires, there are nearly 170,000 car fires a year[1] and 253 million cars[2] on US roads; that's a rate of 1.5 in every 1,000 *every year the car is on the road*, but Tesla's at 0.06 in 1,000 or 25 times less likely for the whole three year period. Even better when you consider that the majority of Teslas are exported.

[1]: http://www.nfpa.org/research/reports-and-statistics/vehicle-fires/highway-vehicle-fires
[2]: https://en.wikipedia.org/wiki/Passenger_vehicles_in_the_United_States

[Rick Law]

Are Electric Vehicles a Fire Hazard?
Lithium-ion batteries have risks, but they can be managed to prevent fires in EVs.
Read the article ; http://www.technologyreview.com/news/521976/are-electric-vehicles-a-fire-hazard/

Erik, your other reply with the data (on vehicle fire stats) is interesting.

This particular article you cited, I was reading it with interest until I conclude it is probably not "balanced" reporting/writing.  It seem to be excuse making to pacify fear of fire rather than real reporting.  The passage that particular strikes me as it being an opinion piece rather than a technology report is:
"In two cases, the cars ran over large metal objects at highway speed; the third car hit a concrete wall. No one was hurt: a warning system allowed the drivers to pull the car over and get out before smoke started coming from the battery pack, and the design of the battery pack slowed the spread of the fire, which never made it into the passenger compartments..."

If after hitting a concrete wall and the people in the car can just "pull over and get out".  To me, it is not a credit but an indictment that the car caught fire in a minor accident.

I click the author's name, and I see his phrasing in his mini-biography.  He is not hiding his bias.
"Growing up, I lived for a time in the Philippines, where I knew people who lit their tiny homes with single lantern batteries or struggled to breathe through the dense diesel fumes of Manila, so I have a feel for the pressing need around the world for both cheap energy and clean energy."

Quote:"Lit their tiny homes with single lantern battery" - battery?  You have to be kidding me.  This is like saying the poor folks have low quality silk shirts which already lost it's sheen.  Don't tell me, instead of buying their "single battery", they troll the garbage for used battery?  This guy have not seen how really poor people live.

Of all the needs of people in places like Philippines and other poor areas, light is about the last thing the "bottom-of-the-economy" people needs.  From food to clothing to just a shelter from the elements.  Those needs he didn't see.  He saw the needs of providing clean energy for the lantern that most of those couldn't afford, clean or otherwise.

Probably another too-well-to-do son of an ex-pat...

Clean energy has its place and has a role to play in this world, but poor areas an't it.  For those on the edge of existence, when making to tomorrow is already questionable, giving them a cleaner future is meaningless.

[AlxDroidDev]

Looking at the bright side, this driver won't have to plow snow from his driveway.

[tom66]

Regarding the Mexico incident.

It's a testament to the car's design that you can walk away from this:


Notice, the passenger cell is intact. The rest of the car is seriously damaged but the crumple zones did their job.

A fire began approximately 15 minutes after this incident, probably due to damaged coolant lines and hot parts, or small amount of damage to the battery which the coolant system could not recover from. Incidentally, in the case of a serious battery malfunction (overtemperature) the coolant system goes into over-drive, and will completely drain the HV pack trying to cool it. Tesla has a patent on some software algorithms for this system.

Incidentally, the video of the fire shows that while the whole front end of the car is on fire, the battery is still functioning (locking/unlocking can be heard, charge port light is lit up and a high-voltage circuit fails.  (At least for a few minutes.)


In the case of the LA crash, the dashboard still worked & was flashing up error messages about the car needing service. Well, it wasn't wrong.

[Rick Law]

Back to the serious side.  Not having done an actual count, Tesla fire seem to be occurring at about one per 3 to 6 months.

This doesn't follow.

AFAIK there have been seven Model S fires, but it's been released for just over three years now
...

What "doesn't follow"?  (Not being argumentative, a real question)

I put "seem" in bold, it seems to be 3 to 6 month each.  I am recalling the frequency I read about such fire.  But my recollection matches the 7 in 3 years you cited.  (36/7 = 5.8 months) rounds to one every 6 months.

[Rick Law]

Regarding the Mexico incident.

It's a testament to the car's design that you can walk away from this:


Notice, the passenger cell is intact. The rest of the car is seriously damaged but the crumple zones did their job.
...
...

It is not survival of the driver that I drew the conclusion of the other article decsribed a minor accident.  While this car caught fire like the Telsa the author described, this accident may not be comparable.  I don't think this driver "pull to the side" after "a warning system" alerted him/her.

To be able to "pull to the side" implies control of the car, and the car being movable.  The picture you inserted shows a car that can no longer be controlled nor "pull to the side".

[Rick Law]

Quick math drill: what’s the difference between 15,000 divided by four and 117,000 divided by 5,000? 

The answer: it's the difference between the odds your plug-in car will be involved in a fire that may or may not be its fault (one in 3,750), versus the odds of a gas station somewhere in America bursting into flames every year (one in 23).

Source. http://www.greentechmedia.com/articles/read/ev-related-fires-ignite-media-firestorm-gas-stations-burn-ignored1

This then in all likelihood is the same physiologic issue of comparing danger of "driving to the airport" vs "airplane accident".

Take your numbers at face value, it is similar to "airplane accident" is significantly less likely to occur then the "drive to the airport."  The "airplane accident" appears far more threatening and far less survivable over "accident on the way to the airport".

[Kjelt]

Yepp..
U.S. fire departments responded to an estimated average of 152,300 automobile fires per year in 2006-2010. These fires caused an average of 209 civilian deaths, 764 civilian injuries, and $536 million in direct property damage.

Nope! You should only report the same accident situation.
This car was brand new powered down standing still on the parkinglot.
So lookup how many NEW petrol fueled cars burned down from spontaneous selfcombustion. That is the case here, stick to the facts.
Then for proper statitics calculate the percentage on the total population of new fuel cars.

[station240]

http://www.teslamotorsclub.com/showthread.php/24797-BEHIND-THE-BUILD-Project-LBR-Tesla-Model-S-Bespoke-Make-Over

Photo #5 in this thread shows the back of the charger port (orange thing), and the red cables that feed back to the electronics under the floor/back seat. Fire started in the rear (see photo on page 2 of this thread). Only possible causes are the charger port itself, or the electronics.

As for the melted car and the fire itself, alloy wheels (aluminium) commonly melt during fires. Also I expect the rear tires caught alight. 

[tom66]

It is not survival of the driver that I drew the conclusion of the other article decsribed a minor accident.  While this car caught fire like the Telsa the author described, this accident may not be comparable.  I don't think this driver "pull to the side" after "a warning system" alerted him/her.

To be able to "pull to the side" implies control of the car, and the car being movable.  The picture you inserted shows a car that can no longer be controlled nor "pull to the side".

You're confusing the road debris incidents. In the road debris incidents objects struck the underside of the battery pack and punctured the sheet steel metal protecting the modules. A short circuit or fire occurred due to damaged cells and/or ruptured coolant systems.

In both cases, as has been documented on the internet, the car continued operation after the incident and had a graceful failure. The system first noticed the isolation fault, so the message: "Car May Not Restart" appeared, indicating the car would continue operating but would not restart as the isolation fault needs to be resolved. This is probably caused by leaky adjacent cell bank readings - possibly due to leaking coolant. Next, "Battery Coolant System Fault" (or similar) appeared with power being limited to roughly 50kW (sufficient for highway cruise but makes the car slow to accelerate.) After some time, temperature sensors began reporting anomalous readings and the car finally clicked the pack contactors over to OFF (with message "Pull Over Safely"), which would give the driver sufficient time to coast the car in neutral and stop the car. 

After the car shut down and the drivers exited the vehicle, a fire occurred in the pack. As the coolant system was damaged, the car could not cool the battery, and the fire spread amongst modules. The battery vents through the front wheel wells; eventually, the fire and heat causes the rest of the car to burn.  The car is wrecked, but the firewalls keep the cabin away from the fire for as long as possible. In fact, in the Tennessee case the fire was contained well enough that the car would still lock and unlock properly, and reportedly "the interior was untouched" (https://www.teslamotors.com/en_GB/customer-stories/model-s-owner-tennessee)  Would you believe this car has functioning electrical systems? http://bizbeatblog.dallasnews.com/files/2013/11/Tesla-model-s-fire-front.jpg http://www.gannett-cdn.com/-mm-/177c43cbf58e55f4d2dfb1862bd86afc2af6dfb9/c=0-345-1628-1571&r=x408&c=540x405/local/-/media/USATODAY/USATODAY/2013/11/07/1383850858000-AP-TESLA-FIRE-59680562.JPG

The firemen arrived promptly and applied water to the flames. They were about to pry open the doors, so I pressed my key button and the handles presented and everything worked even though the front of the car was on fire. No flames ever reached the cabin, and nothing inside was damaged. I was even able to get my papers and pens out of the glove compartment.

That shows that the firewalling between pack and vehicle systems was so good that the intense fire didn't even disrupt the standard 12V automotive functions of the car, including the battery which is adjacent to the right hand wheel well.

As for the Mexico case I'm not sure it really matters if the car tells you to pull over safely after it's essentially destroyed. You can make that decision yourself. And it doesn't matter too much if the car burns 15 minutes after the accident occurs. For what it's worth that car went through a concrete wall, over a roundabout hump and into the crash barrier before coming to a stop. The battery fire is the least of the concerns.

I think you really need to look at the different types of incidents that occur. If a petrol car gets into a serious accident, it's actually quite possible you'll get an explosion, though not in the Hollywood style:


Whereas such an explosion is impossible with a li-ion electric car because there is no highly volatile fluids and hot engine parts to mix. If you rupture the pack, the worst you'll get is a slow burning fire that takes over an hour to reach the passenger compartment and is relatively safely contained in the front of the car. It would be possible to get out of the car in time, or for emergency services to respond and rescue trapped occupants.  Obviously, no one wants a fire, but when you're carrying around megajoules of energy in a small package, fire is a possible outcome after serious trauma to such an energy storage device. The engineering challenge then becomes limiting the rate at which the fire will spread within the battery pack and keeping the fire away from the rest of the car for as long as practically possible.

[Rick Law]

It is not survival of the driver that I drew the conclusion of the other article decsribed a minor accident.  While this car caught fire like the Telsa the author described, this accident may not be comparable.  I don't think this driver "pull to the side" after "a warning system" alerted him/her.

To be able to "pull to the side" implies control of the car, and the car being movable.  The picture you inserted shows a car that can no longer be controlled nor "pull to the side".

You're confusing the road debris incidents. In the road debris incidents objects struck the underside of the battery pack and punctured the sheet steel metal protecting the modules. A short circuit or fire occurred due to damaged cells and/or ruptured coolant systems.
...
...

If there was a confusion, it was introduced by the article's author.  The question perhaps is whether such confusion was by intend (pushing an opinion) or not.  My original quote from his article was:

"In two cases, the cars ran over large metal objects at highway speed; the third car hit a concrete wall. No one was hurt: a warning system allowed the drivers to pull the car over and get out before smoke started coming from the battery pack, and the design of the battery pack slowed the spread of the fire, which never made it into the passenger compartments..."

I quoted that verbatim.  Note his use of semi-colon and colon right there.  That the "pull the car over" applies to the car that hits the concrete wall is exactly what he said.  With the semi-colon, it is arguable whether it applies to the first two car that hit the large metal objects, but that two car is not what is in discussion here.
   
The significant words were: "Third car hit a concrete wall....a warning system allow the driver to pull the car over..."  Those are his words.  So clearly after the accident the driver still can control the car and the car was mobile.

The car you show was in a significantly worst accident.  The car in your picture is without a front wheels and with at least another wheel dislodged; and the car was sitting on a road divider.  Clearly the car in your picture is no longer mobile nor was the driver able to exert any further control of the car after the accident.  So while your pictured car also caught fire, it was not a comparable accident -- at least not comparable to the accident this author described in his article.

[tom66]

The car in the picture is the car that was involved in the fire in Mexico. That was the second Model S fire occurring only weeks after the first fire. That car slammed through a concrete barrier, shot across a roundabout/traffic island and collided into a crash barrier. Reportedly, according to police, the driver was drunk and travelling in excess of 100mph for some time (though probably not at the collision point.)

 You will hear no argument from me that the car is no longer drivable! I think the author had meant to refer to the two cars that were victims of road debris as both of those vehicles were still drivable for some time after the incident (several minutes) and so the "limp mode" gave them sufficient time to pull over safely and exit the vehicle. The car in Mexico caught fire only after the collision.

[Rick Law]

The car in the picture is the car that was involved in the fire in Mexico. That was the second Model S fire occurring only weeks after the first fire. That car slammed through a concrete barrier, shot across a roundabout/traffic island and collided into a crash barrier. Reportedly, according to police, the driver was drunk and travelling in excess of 100mph for some time (though probably not at the collision point.)

 You will hear no argument from me that the car is no longer drivable! I think the author had meant to refer to the two cars that were victims of road debris as both of those vehicles were still drivable for some time after the incident (several minutes) and so the "limp mode" gave them sufficient time to pull over safely and exit the vehicle. The car in Mexico caught fire only after the collision.

Then we are in not in disagreement.  The author did not describe the situation clearly.  He described an accident where the driver had control after the accident and can "pull over".  Your picture clearly show a car that isn't able to go anywhere.

While the confusing passage is by intend or not really doesn't matter.  Readers form opinion about about him either way.  In my view, it was because he was pushing an opinion.  Inferring such, I click the link for his mini-biography and from that, drew my own conclusion.

Lets get back to the main topic...  Tesla, a good place to BBQ or good transportation.

[EEVblog]

Yepp..
U.S. fire departments responded to an estimated average of 152,300 automobile fires per year in 2006-2010. These fires caused an average of 209 civilian deaths, 764 civilian injuries, and $536 million in direct property damage.

Nope! You should only report the same accident situation.
This car was brand new powered down standing still on the parkinglot.
So lookup how many NEW petrol fueled cars burned down from spontaneous selfcombustion. That is the case here, stick to the facts.
Then for proper statitics calculate the percentage on the total population of new fuel cars.

Err, no, the Tesla was charging at the time was it not?
If so, then it's fair to compare fire when the Tesla is charging, and when petrol cars are being filled at the service station.

[helius]

Amphiboly is a common symptom of an invalid conclusion being pushed, so it's no surprise that it shows up in that article.

[Poe]

[Kjelt]

Err, no, the Tesla was charging at the time was it not?
If so, then it's fair to compare fire when the Tesla is charging, and when petrol cars are being filled at the service station.

Uh I guess analogies between electrical and fuel cars are not easy to make in this case, since their individual offroad weak spots seem to differ too much.

[tom66]

An EV charges for longer than a petrol vehicle, but a petrol vehicle has higher volatility during the refuel time. An EV can catch fire due to battery fault, overheating components etc. A petrol vehicle can catch fire if there is a spark or hot components near fumes.

[janoc]

Quick math drill: what’s the difference between 15,000 divided by four and 117,000 divided by 5,000? 

The answer: it's the difference between the odds your plug-in car will be involved in a fire that may or may not be its fault (one in 3,750), versus the odds of a gas station somewhere in America bursting into flames every year (one in 23).

Source. http://www.greentechmedia.com/articles/read/ev-related-fires-ignite-media-firestorm-gas-stations-burn-ignored1

I suggest you learn something about statistics first if you want to make such comparison.  Hint - how many are there electric plug-in cars vs gasoline cars using those gas stations that catch fire? Likely few orders of magnitude.

So it is quite obvious that you will have more accidents at gas stations and fires (electric car won't catch fire because of a moron checking the amount of gas in the tank with a lighter ...), simply because of their much larger numbers.

It is a completely meaningless comparison.

[helius]

Well, the comparison was phrased as a rate, so the impression is that it already accounts for popularity. But the bigger problem is that many of the gas station fires are from predictable causes like people smoking or collisions with the pump; the Tesla fires being discussed weren't predicted and their cause is uncertain.
To expand that point a bit, when we can predict a base rate of accidents at gasoline pumps, we can reason about how much damage they cause and whether things need to be changed to reduce the damage. If an EV catches fire for unknown reasons, we can't reason about how big a problem it is: simply counting the rate of accidents is no good for extrapolation; what if it was caused by software? If the cause is an unknown software fault they could ALL catch fire tomorrow.

[Kjelt]

yeah and what about 10yrs and older cars where the deteriorating petrol-tubing might be the cause of fire or to mention the enormous amounts of cars that are set at fire at newyears night at purpose or through fireworks, etc. etc.?

[Mechanical Menace]

yeah and what about 10yrs and older cars where the deteriorating petrol-tubing might be the cause of fire

Isn't there an annual Dutch road worthiness check? Even the UKs MOT, which in some respects is quite useless, would pick up on that before it was a problem.

or to mention the enormous amounts of cars that are set at fire at newyears night at purpose or through fireworks, etc. etc.?

Is that really something that happens en masse in the Netherlands?