Print 76 comment(s) - last by Totally.. on Nov 16 at 8:07 PM

This is the third fire in under a two-month period

The National Highway Traffic Safety Administration (NHTSA) has decided to look into the latest fire involving Tesla Motors' Model S. 

According to The Detroit News, NHTSA will review what is now the third Model S to catch fire in under a two-month period. This most recent incident took place near Smyrna, Tennessee. 

The Tennessee fire took place when a Model S driver hit a tow hitch on Interstate 24, which damaged the car’s undercarriage and caused a fire. The driver was able to exit the vehicle safely.

“NHTSA is in close communication with Tesla and local authorities gathering information about the incident to determine if additional action is necessary,” said NHTSA.

Special focus is being placed on the fact that the Model S' battery is located near the underside, making the battery an easy target when striking debris or hitting pavement -- thus increasing the risk of a fire.

This could lead to stronger methods of protecting the EV's battery pack. 

NHTSA has not opened a formal investigation on the Tesla crashes yet, likely because this is the first incident it's addressing.

Model S fire in Smyrna, Tenn. [Image Source: Associated Press]

But this isn't the first Tesla fire to occur. In early October, a Model S driver in Kent, Washington was traveling southbound on state Route 167 when he hit a piece of metal debris on the freeway. He then exited the freeway, and the car became disabled right before he smelled something burning. The car caught fire.
Tesla spokeswoman Liz Jarvis-Shean said the fire was caused by a large metallic object hitting one of the battery pack’s modules. NHTSA did not investigate the fire at the time because of the partial government shutdown, which suspended such activity. 

Another Model S fire occurred shortly after in Mexico, but that's out of NHTSA's jurisdiction. 

Tesla said all three fires were caused by crashes; not spontaneous events.

Tesla and its Model S have been in the spotlight a lot this year after the company successfully paid off its government loans nine years early, pulled a profit, unveiled new tech for its electric car and the Model S even snagged the highest safety rating from the NHTSA. But it's unclear if this hiccup will further heighten the fear surrounding lithium ion batteries for cars, and possibly even take a toll on Model S sales. 

Source: The Detroit News

Comments     Threshold

This article is over a month old, voting and posting comments is disabled

RE: Lithium batteries burn
By Motoman on 11/11/2013 12:32:29 PM , Rating: 3
Oh, actually that's been posted already. A few posts up from here someone posted that there's ~150,000 car fires per year in the US.

If the ~38,000 per month were true, that figure would be more like 458,000. More than 3 times the actual rate.

Ergo, it would appear that the Tesla is roughly 3 times more likely to catch fire than the average vehicle.

RE: Lithium batteries burn
By daboom06 on 11/11/2013 12:54:39 PM , Rating: 5
measurements from small sample sizes do not produce meaningful statistics. we're looking for the regime where n ln(n) reasonably approximates ln(n!). and that doesn't happen until around 50.

RE: Lithium batteries burn
By daboom06 on 11/11/2013 12:55:41 PM , Rating: 2
sorry, the approximation is n ln(n) - n

RE: Lithium batteries burn
By Motoman on 11/11/2013 1:03:56 PM , Rating: 2
3 isn't the sample size - 20,000 is. The rate is 3 fires per 20,000 units per month.

RE: Lithium batteries burn
By daboom06 on 11/11/2013 2:32:01 PM , Rating: 5
the sample size is small if the number of positive events is 3.

RE: Lithium batteries burn
By Solandri on 11/12/2013 5:13:48 PM , Rating: 4
Actually he's correct that the sample size is 20,000 here.

The problem with having only 3 incidents is that the uncertainty becomes larger relative to the incident rate as the incident rate becomes very small.

z = 1.96 for a 95% confidence interval
std_err = sqrt[ (p)*(1-p) / n ]
margin of error = z * std_err

e.g. If there were 10,000 events (50% observed incident rate), then the margin of error is:

1.96 * sqrt( (.5)*(1-.5)/20000 ) = 0.006929646

So the true incident rate is 50% +/- 0.69%

But if you calculate it for just 3 incidents (0.015% observed incident rate), you get:

1.96 * sqrt( (.00015)*(1-.00015)/20000 ) = 0.000169728

So the true incident rate is 0.015% +/- 0.017%. In other words, your margin of error is bigger than your incident rate. With a 95% confidence interval, all you know is that the true incident rate lies somewhere between 0 per 20,000 and 6.8 per 20,000. Which pretty much is the same thing as saying you don't know what the true incident rate is, other than it being a very small number.

RE: Lithium batteries burn
By Keeir on 11/11/2013 1:03:52 PM , Rating: 2
Mmmm... I've heard about 3 reports of Teslas catching fire.

That includes both Roadsters and Model Ss.

Roadsters have ~3 years on the road for a rough average of 6,000 auto-years

Model Ss have accumulated around 10,000 auto-years.

3/16,000 auto years ~ 2 per 10,000 auto years (rounded up)

There are roughly 150,000 auto fire a year per ~250 million cars or ~6 per 10,000 auto years.

Seems we have a ways to go. However, many of those 250 million cars are significantly older and potentially poorly maintained. All Teslas should be new and have no significant maintainence issues.

I'd say that the need to add a more substaintial skid plate/energy absorber for the front undercarriage. Several Model Ss and Roadsters have crashed without fire. Simple fix.

Your use of a 1 month sample, monthly rate is silly. For months Tesla had a "zero" rate.

RE: Lithium batteries burn
By kleinma on 11/11/2013 1:27:46 PM , Rating: 2
How many of the 150k vehicle fires are due to a crash? The 3 tesla models were all involved in some sort of collision with a foreign object, versus being a malfunction or defect in the engine/batteries.

So extrapolate so you are comparing fires of only vehicles that were in a collision, and then where are the figures?

RE: Lithium batteries burn
By ilt24 on 11/11/2013 1:34:22 PM , Rating: 3
After the first fire Musk wrote this:

"The nationwide driving statistics make this very clear: there are 150,000 car fires per year according to the National Fire Protection Association, and Americans drive about 3 trillion miles per year according to the Department of Transportation. That equates to 1 vehicle fire for every 20 million miles driven, compared to 1 fire in over 100 million miles for Tesla. This means you are 5 times more likely to experience a fire in a conventional gasoline car than a Tesla!"

So now with 3 fires his metric would show 1 fire per 33M miles for Tesla vs 1 fire per 20M miles for all vehicles.

Not sure how useful this or your metric is; also one of the three Tesla fires was in Mexico which throws off the numbers.

RE: Lithium batteries burn
By tayb on 11/11/2013 1:48:24 PM , Rating: 4
I would say at these sample sizes the numbers are completely meaningless. 3 fires, 20,000 vehicles sold, and only 100 million miles driving.

There just isn't enough data to make reasonable conclusions right now.

RE: Lithium batteries burn
By Totally on 11/11/2013 3:29:30 PM , Rating: 1
It was enough for Tesla to issue a statement saying the it's vehicles are statistically less prone than ICE powered vehicles.

RE: Lithium batteries burn
By lelias2k on 11/12/2013 8:52:05 AM , Rating: 2
Because they are in a very fragile position.

They are going against some of the most powerful corporations in the world (oil), which are fighting left and right to prove that their technology is the wrong choice (I wonder why).

Although it is a fantastic car according to pretty much everybody who drove it (I haven't read or heard otherwise so far), it still has to prove itself in order to win the masses when the reasonably affordable models come to the market.

If the company fails now because of unjustified widespread panic (there's not enough sample to justify it and nobody got hurt in a Tesla yet, AFAIK), we might never see what they would be able to do in a few years with better and cheaper technology.

RE: Lithium batteries burn
By ven1ger on 11/12/2013 7:23:43 PM , Rating: 3
This is a key factor that everyone posting stats on vehicles.

The Tesla and other similar cars are in its infancy, so problems will arise. Gas powered cars have been on the road far longer that I have been alive, yet, we are comparing a mature gas powered vehicle and the amount of fires it has versus a virtually new type of car and the number of fires. Probably need to go back to how many gas powered cars caught on fire the first time it rolled out.

I think that as these kinds of cars mature, you'll find they are safer as they realize through real world testing that certain fragile parts should not be placed in certain areas or they need to beef up the shielding in some areas. Even the gas powered automobile had to go through all of this real world testing and we have much much more reliable vehicles. Same goes for these sort of vehicles.

RE: Lithium batteries burn
By Totally on 11/16/2013 8:07:10 PM , Rating: 2
I wouldn't ever dare say that the technology behind Tesla's vehicles is virtually new, batteries and dc motors have practically been around since the discovery of electricity, and outside of that is existing and continuing automotive tech, we've just about had a battery powered everything except planes and cars. When when the first automobiles rolled out there weren't computer simulations, extensive crash safety test, or safety regulations in general. Even though battery powered vehicles are 'new' the tech behind them is mature just not pushed to the limits like ICE vehicles. These problems are typical QA issues any new company in any field experiences with a new product. We don't hold back criticism for them, so why should we for Tesla motors. That said it's too early to say anything FOR or AGAINST, which is what i was trying to bring up earlier that it was hypocritical that the available data can be used in defense but unacceptable grounds criticism.

RE: Lithium batteries burn
By hrrmph on 11/11/2013 2:03:21 PM , Rating: 2

After taking into account the latest fire (the 3rd fire), here are the statistics derived from Elon Musk's previous statements on USA based fires:

All automobiles: 1 fire every ~20 million miles (or about 17 fires every hour).

Tesla S: 1 fire every ~35 million miles (throwing in the fairly violent Mexico accident so as to not ignore it, and rounding up the miles a bit to reflect continued safe operation of the rest of the fleet, thus giving us easy to look at numbers).


Now I think the real question should be "Are Elon Musk's statistics accurate?"

I dunno, but this topic sure took our minds off of Boeing's batteries, didn't it??

RE: Lithium batteries burn
By Reclaimer77 on 11/11/2013 6:56:16 PM , Rating: 3
Then maybe Musk can cook up some "statistics" to explain why there are far more Nissan Leaf's on the road, and not a single one has caught fire?

RE: Lithium batteries burn
By AssBall on 11/13/2013 5:49:51 AM , Rating: 2
Ha... he can call it "cooking" with Lithium If a Nissan Leaf catches fire in the forest and no journalists are there, does anyone care?

"Spreading the rumors, it's very easy because the people who write about Apple want that story, and you can claim its credible because you spoke to someone at Apple." -- Investment guru Jim Cramer

Most Popular ArticlesAre you ready for this ? HyperDrive Aircraft
September 24, 2016, 9:29 AM
Leaked – Samsung S8 is a Dream and a Dream 2
September 25, 2016, 8:00 AM
Yahoo Hacked - Change Your Passwords and Security Info ASAP!
September 23, 2016, 5:45 AM
A is for Apples
September 23, 2016, 5:32 AM
Walmart may get "Robot Shopping Carts?"
September 17, 2016, 6:01 AM

Copyright 2016 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki