Hacker News new | past | comments | ask | show | jobs | submit login

We're reading this on Tesla's site, and they obviously wouldn't have posted it if it weren't at least somewhat flattering, so we have that lens to view it through.

But, the letter does raise the important idea that before we decry how awful the failure mode is ("the car caught on fire!"), it's worth considering what the alternative would be---what happens in similar situations in combustion engines?




> We're reading this on Tesla's site, and they obviously wouldn't have posted it if it weren't at least somewhat flattering, so we have that lens to view it through.

In this case, we've seen reports from the driver in 2 of the 3 accidents, so it's hard to make a case for selection bias. Though feel free to argue it further if you know details of the crash in Mexico that I'm not aware of.

Driving a car is one of the least safe things we do regularly, FWIW. With the population large enough, we are virtually guaranteed to see more of these.

Unfortunately, the media does terrible things to our risk perception by giving us incomplete evidence and leaving us to extrapolate. For example, to compare whether the Tesla is unsafe--and for all we know, maybe there is some flaw that helps road debris catch on the plate and puncture it--it would be more informative to see fires/#miles driven for a large cross-section of common cars, as well as fatalities/#miles driven and the like. Of course, even that is somewhat biased, because there might be effects where, hypothetically, risky drivers are drawn certain cars or something like that.


Nothing. Conventional cars are designed not to catch fire when they run over road debris, especially not cars built within the last two decades. At worst, debris may cause a flat if it punctures one of the tires.


>At worst, debris may cause a flat if it punctures one of the tires.

That's dismissive hyperbole. Debris regularly causes serious damages to all vehicle types. If the object in question, as in this case, is hard enough and hit with enough force to penetrate the battery, it's hard enough and hit with enough force to seriously damage any system it hits. It may not crack the engine block, but it can destroy radiators/suspension/brakes/fuel tanks/etc. The worst case scenario is most certainly not "a flat".


>At worst, debris may cause a flat if it punctures one of the tires.

That's just not true. I had about $2k in damage on a Mini Cooper (2006, stock) that hit the remnants of a blown out semi truck tire in the road. Similar situation to the blog post -- a pickup truck cleared it but I did not. I was not in a safe position to swerve.

The tire chewed up the lower front bumper and punctured the radiator. I had to immediately pull over and have the car towed. It happened to hit below the foam & all the fancy crash reinforcement on the bumper. Good times.

That was just 70lbs of steel-wrapped rubber. I can easily imagine a trailer hitch as described taking out a radiator or oil pan on a modern car. Unlikely to cause a fire for sure, but it'll certainly disable a vehicle.


https://en.wikipedia.org/wiki/Road_debris

"In 2004, a AAA Foundation for Traffic Safety study revealed that vehicle-related road debris caused 25,000 accidents—and nearly 100 deaths—each year."


The term "road debris" can mean anything from dust to entire engine blocks[1]. The debris in question here is a metal tow hitch.

http://en.wikipedia.org/wiki/Road_debris#Examples

See my other comment upthread for an example of a gasoline car catching fire because of cardboard boxes.


Normal vehicles don't have their fuel tank along the whole bottom of the car, let alone that far to the front. All that is for a reason.

Tesla made a choice to arrange packs like this for swap purposes. It certainly wasn't going all out for safety. Throw in the fact battery tech really isn't all there yet and you get a very large battery that compromises have to be made.

Tesla chose a compromise based on their swap tech. They could have put more batteries in the rear or made the stack shorter. Used a central tunnel for more, etc.


>Tesla made a choice to arrange packs like this for swap purposes.

That might have contributed, but I think lowering the center of gravity was the main consideration.


A normal car would not tell you there was a problem and suggest corrective action, then demand you pull over when safety parameters were exceeded, keeping you safe and healthy with minutes to spare.

The first you would know there was a problem in a normal car, would be when you felt or saw the fire.


ummh... you really think Tesla's PR team didn't write that and ask the guy to sign it? implying how nicely they were going to treat him if he did? Sure, it's quite possible as a fanboy he did it totally on his own. But I wouldn't bet that way, and I have to think most people who know how the tech industry works would feel the same.


I'm guessing most Tesla owners at this point are rich, well-educated and level-headed people who are also early adopters and a fan of the car and company. It seems reasonably likely that it's genuine.


I find it unlikely that Tesla PR would attribute their cars' safety to a supernatural being.


ha, good point.

not sure if it adds to the overall credibility.

bottom line, the guy likes the car enough to write/cooperate, even after it hit a bump and burned up, how much of that is the car's greatness, rose-colored fanboi glasses, good old-fashioned working the media...unclear.


When was the last time you saw a car accident on the news where the car burst into flames ?

Maybe it's just a case that combustion engine based cars have had more time to resolve this issue properly.


>>When was the last time you saw a car accident on the news where the car burst into flames ?

Last month, right here in Seattle: http://www.king5.com/traffic/news/Collision-causes-car-fire-...



Local traffic reporters (in Atlanta at least) have been known to call cars catching on fire 'Car-B-Qs'. It's a distasteful name but indicates that cars catching on fire isn't that uncommon.


it's worth considering what the alternative would be---what happens in similar situations in combustion engines?

What happens then? There are several innuendos throughout this thread that allude to some sort of catastrophic scenario if it were a normal vehicle, however this seems unsupported.

Yes, normal combustion engines catch on fire, but they don't explode in a fireball when they hit debris on the highway (there is absolutely nothing rare about hitting debris on the highway).

Further there are millions of combustion engine vehicles on the roadway -- there are currently some 250 million traditional vehicles in the United States. Last I read Tesla is delivering something in the range of 30,000 vehicles per year.


> Yes, normal combustion engines catch on fire, but they don't explode in a fireball when they hit debris on the highway

To be fair, none of the debris I've ever seen or hit on the highway would have made holes in a quarter inch steel plate, and I have driven the better part of 25k miles/year commute over some very busy highways for approximately the past decade, so I've seen such random debris as chairs & couches in the middle of the road in addition to the usual assorted vehicle parts.

There is also a car fire roughly every 3 minutes, just to keep things in perspective.

I say that as we are guaranteed to see more fires given enough time. With them averaging out to one every 3.x minutes, it's only a matter of time before we go through this again.


> There is also a car fire roughly every 3 minutes, just to keep things in perspective.

Are you joking ? How does that keep anything in perspective ?

There are over 1b cars in the world and 50k Teslas. That's 0.005% of the total car population.


> How does that keep anything in perspective ?

The Tesla has been available for 140 days, as of the time I'm posting this and using the June 22nd release date[1]. So my back-of-the-napkin estimates say there have been about 63,000 car fires in that time so that 0.005% of the car population is responsible for 0.000048% of the car fires in the past 140 days, which does not seem disproportionate. There are other, even better ways to compare the risk for which I do not have data available. Feel free to research your own numbers and compare.

[1] http://www.teslamotors.com/model-s-has-arrived


But remember we are talking about an expensive, new car with a lot of modern safety features. Which is the complete opposite of most of the cars you see around the world. My point was that comparing Tesla's record with existing cars will never make sense since we simply don't have all of the facts.

It seems obvious though that Tesla being a very different design will have different incident characteristics.


June 22, 2012.


Oops. That makes things even more ridiculous, if there have been 3 battery incidents in all this time. Put 505 days into my calculation instead of 140...

As to the point about them all being new cars, well, new cars are safer. If safety is what we're after then yes, a modern car is a good thing. If you're saying that the safety will decay as they age, well, that's reasonable but only time will allow us to correct our calculations there.

That might mean that we need more time to make a full determinations, but that's not a bad thing. This tells us about their current safety which is what's relevant now.


That leaves approximately 227,250 fires since the Tesla has been introduced, of which Teslas are known to be responsible for three. Or 0.000013% of all car fires in the relevant time.

That said, the other test might be more relevant, as there were more actual Teslas on the road in the past 140 days, rather than at day zero of launch. I don't have the figures to calculate, say, fires/#miles driven.


I'd point out that those innuendos are based on manipulative statistics: the standard is comparing the rate of car fires, but most of those 250 million cars are far older than Tesla Model S (decades as opposed to months old). A proper apples-to-apples comparison would look at new cars (within a year, lets say) and compare those rates.


Simple calculations show that the Teslas are roughly 25x more likely to catch fire in a collision, but I suspect they are much less likely to catch fire for other reasons. So just don't hit anything.


What's the error rate on that? I don't think we've seen enough Teslas involved in major accidents to even begin to determine how likely they are to catch fire compared to other cars.


You don't need to know the accident rate--you just need to know the rates of fire from an accident per car-year driven. Tesla doesn't have a long history here, but they do have 10K cars on the road.


I know nothing about the physics of it but would that not depend on the causes of fires? I may be misunderstanding but it sounds like you're implying that fires are more likely in older cars - maybe that is actually the case (I'd be willing to believe it).


A huge number of vehicle fires are electrical in nature.

I've been a firefighter for ~10 years. Nearly every vehicle fire I've seen falls into one of three categories:

1) Deliberately set 2) Faulty/old/corroded wiring (by far the most common) 3) Some flammable material coming into contact with something really hot (commonly after an accident where a car has gone off the road into tall dry grass)

Fires due to the fuel tank failing are _extremely_ rare (I've never seen one, or even heard of one in neighboring areas).

It may be related to the fact that I'm in an area that uses lots of nasty stuff to deice roads, but corrosion is a major issue with vehicle electric system, and the older a vehicle is, the more likely it is to experience a 'spontaneous' fire.


Another important effect is that some cars have known defects putting them at increased risk of fire, and some of those cars are still on the road, which inflates the statistics for the average automobile.


The Tesla didn't explode in a fireball, it vented the heat to the front, away from the battery pack and the cabin, and then the front caught fire.


I am not implying that it did explode in a fireball. We see the evidence several times now that it was a controlled, escapable loss of property, which is an ideal worst-case. I was replying to the "what could have happened if" innuendo about combustion engines.


Exactly. The tesla cars are just too delicate. You definitely can't have one without warranty. And I say this as an Elon Musk admirer.


The same is true for other cars in its class, such as a Mercedes S-class, or a BMW 7-series.


> Yes, normal combustion engines catch on fire, but they don't explode in a fireball when they hit debris on the highway

And neither did this one, so I'm not sure what your point is.


there are currently some 250 million traditional vehicles in the United States

And 194,000 vehicle fires.


>Yes, normal combustion engines catch on fire, but they don't explode in a fireball when they hit debris on the highway (there is absolutely nothing rare about hitting debris on the highway).

That is not true. Here's a case of a car catching fire after being hit by cardboard boxes.

http://www.ksl.com/?sid=25850334

The Tesla in the story was hit by a tow hitch looks like this. http://1.bp.blogspot.com/-A22JVrgZjqY/TzKFRNPUN1I/AAAAAAAADh...

I find it very surprising that you think it's common for cars to run over things like that on the road and keep going. I or the people I know have never driven over something like that.

From Wikipedia:

>Road debris is a hazard[5] that can cause fishtailing and damage like a flat tire or even a traffic accident with injury[6] or death. Road debris can cause loss of control crashes, rollover crashes, or penetration of the passenger compartment by the debris.[1][7]

>Released in early 2013, NHTSA data for 2011 showed over 800 Americans were killed that year in vehicle collisions with road debris. Mississippi, Wyoming, Arkansas, Kentucky and Louisiana were the top five states for these crash deaths to most likely occur. Also in 2011, New York and Massachusetts saw significant increases in road debris-vehicular crash deaths, unlike other big, populated states.[8] In 2004, a AAA Foundation for Traffic Safety study revealed that vehicle-related road debris caused 25,000 accidents—and nearly 100 deaths—each year

http://en.wikipedia.org/wiki/Road_debris#Effects


That news story of the conventional car catching fire after an encounter with road debris is the perfect comparison. It happens, and the same outcome often happens as well.

This is not surprising. It's a little surprising that we've seen so many stories of Teslas hitting road debris, but my hunch is that the problem is caused not by being electric, but by a low bottom clearance designed for drag efficiency, battery space, and low center of gravity.

A gasoline car with such low clearance would encounter similar problems, as they surely do every single day. This is completely normal. The way it affects an electric battery is kinda crappy, but it's not necessarily worse than debris hitting a tank of flammable fuel, just different. A tesla would have gone right over cardboard, for example, and been unharmed, while it might get stuck in the underbody of an ICE car and catch on fire.

Different technology, different problems, but the sample size is still too low to make real conclusions. One thing we do know is that there have been no deaths in a Tesla vehicle yet; that can hardly be said for traditional cars.


A gasoline car with low clearance would NOT have the same problems.

The fuel tank is located higher up near the rear axle. Not alongside the bottom of the car as in a Tesla.


Yes this has been standard design for a good while now. The fuel tank is typically above the rear axle and between the rear seat and the trunk. I don't think there's been a car with the fuel tank right over the road and right behind the rear bumper since the 1970's. The Pinto in particular taught us that was a bad design.


The rear fuel tank low down behind the rear axle is still in some currently produced cars, I'm afraid. http://www.autosafetyexpert.com/defect_fueltank.php


Note: either the battery didn't catch fire in the said tesla (see picture), either the battery is extremely well isolated and the fire exhaust point was in the front of the car.

Note2: how do you think the fuel goes from the fuel tank <REAR> to the <FRONT> motor.

Thats how: http://i.stack.imgur.com/87oFf.png

Oh, theres a tube in there between the REAR fuel tank and the FRONT engine. Damn, didn't break laws of physics.


> That is not true. Here's a case of a car catching fire after being hit by cardboard boxes. > http://www.ksl.com/?sid=25850334

To be fair, that was more of case of a cardboard box catching on fire after being hit by a car...


Probably more likely the cardboard box is dragged under the car a few seconds, heat up and catch on fire.


That is not true. Here's a case of a car catching fire after being hit by cardboard boxes.

Are you seriously saying that the car exploded into a fireball? Because of course it didn't, and in fact the combustion was primarily the incendiary material of the cardboard box, just as it would ignite near high power electric motors. That has zero relevance to this story.

I find it very surprising that you think it's common for cars to run over things like that on the road and keep going.

You entirely misinterpreted my post. I stated, nor implied, absolutely no such thing.




Consider applying for YC's W25 batch! Applications are open till Nov 12.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: