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

I'm glad the driver is safe. However, I'm not sure what posting this letter is supposed to accomplish other than show that the driver is still a Tesla supporter. The remainder of the letter seems to be (potentially) counterproductive in that it describes that the car being on fire wasn't as dangerous as it sounds. Tesla isn't going to be able to tackle a perception problem by promoting safety features of the car when in flames.

It's supposed to show that the Tesla car is structurally sound and prevented the object on the road from ripping further into the car than it otherwise would have. It's also supposed to show that the car got the driver out of a dangerous situation by displaying warnings early. Finally, it is supposed to show that the car is resilient when bad things happen because it is still drivable.

I'm just explaining the viewpoint of the letter since you were unsure. I'm not opining that it is effective/ineffective or productive/counterproductive.

And what would happen if the person was rendered unconscious during an accident ?

I think the point is still valid that the safety features are nice but the core issue is that the car burst into flames during a seemingly minor accident.

Running over a trailer hitch at 70mph is a major accident. A smaller, lighter car could have been thrown off the road or rolled over, in addition to having the object puncture and invade the passenger compartment possibly causing injury, and loss of control. You're right in that it would be unlikely to cause a fire, though I wouldn't really worry about that if I was dead already.

"A smaller, lighter car could have been thrown off the road or rolled over, in addition to having the object puncture and invade the passenger compartment possibly causing injury, and loss of control."

...or maybe not. You're just speculating.

I've run over stuff at high speeds in other cars, and I"m here to tell you that they didn't flip over, puncture, or burst into flames. The most common outcome, in my experience, is that the oil pan or transmission pan gets dented, and starts to leak.

I ran over a metal object on the freeway that punctured the transmission which started pouring transmission fluid onto the exhaust pipe and ignited. The truck was on fire for a few minutes before I was alerted by another driver and pulled over.

Low probability events happen all the time, and I don't think we can draw any solid conclusions from these anecdotes (other than driving any vehicle is potentially dangerous).

> Low probability events happen all the time

Hah! I love how extremely counter-intuitive and completely true this sentence is at the same time.

Cars do burst into flames though. There are around 194,000 vehicle fires on US roads each year from around 250,000,000 registered vehicles, which gives around 1 fire for every 1,300 vehicles. There are so far 3 Tesla fires from around 16,000 Tesla cars on US roads, which gives around 1 fire for every 5,300 of them, so they are doing better than the average. 3 fires though is still far too few to consider this a particularly useful statistic.

In order to determine how well the Tesla S handled these collisions, the stats you cited are useless. We should not count fires caused by types of accidents Telsas have never been involved in. Also, no one is accusing Tesla cars of bursting into flame for no reason, so the cars that are not in accidents aren't important either.

To really understand where the Tesla ranks in safety, we would need a detailed analysis of: the accident, how non-Tesla cars handled similar accidents, how the Tesla is designed to handle such an accident.

I do agree that we have no where near enough data to draw general conclusions about the safety of the Teslas, so I think it's inappropriate to say, "they are doing better[or worse] than the average."

A detailed analysis of "the accident" is precisely how we don't find out how Tesla "ranks in safety".

Once the numbers of cars become large enough we can simply look at the outcome numbers, to determine how likely that outcome is. It doesn't matter if a car is "designed to handle such an accident". What matters is if it does handle a given kind of accident.

A Tesla car is unlikely to ever be involved in a fire at a gasoline station. We probably shouldn't give combustion vehicles a free pass there.

As more and more Tesla cars hit the road, confidence level increases in the stats that are derived. Early data is not meaningless; it can be indicative of later results.

I agree with the rest of the post, and this part might be true of this thread: "Also, no one is accusing Tesla cars of bursting into flame for no reason, so the cars that are not in accidents aren't important either." but not necessarily among the general public.

I think this is precisely the idea they are combating. The initial news reports didn't have any details and just said there was Tesla on fire on the side of the highway.

Correct this a bit and you find 3 per ~100 million miles driven on public highways. that is about ~3x worse than a regular car. 100 million miles is enough data points to be meaningful, but not exhaustive. Also, given these cars are new, the state-space of defects is likely less populated tha for the average car. Historical data show age as a factor in fires, and new cars are less likely than average-age cars to catch fire.

I was surprised that the letter did not go into these types of details. Cars (electric or not) catch on fire. There are statistics on this. The truth about the Model S's safety has not changed. Tesla seems like a more 'sticking to their guns' kind of company rather than a adjusting tactics to handle this situation.

The underside of a Tesla is quite a bit more sturdy than your oil or transmission pans. A piece of debris able to do this would do something far more serious than cause a dent or leak.

The principle difference is that the entire underside of a Tesla is vulnerable (to a first approximation), while a conventional vehicle has a few zones of particular sensitivity (fuel tank, oil pan, gas line, possibly drivetrain).

I'm fairly impressed with Tesla's record so far, but mounting a half ton (literally) of energy storage to the entire underside does raise some interesting questions.

Your anecdotes are not data.

Once I had a rock the size of a golf ball smack into my windshield on the highway. It hit with such force that it made a noise that almost sounded like a gunshot, it was quite startling. It didn't leave even the smallest mark on the windshield though. Another time a tiny pebble the size of a pea hit my windshield and made a very sizeable chip in it.

Anecodtes involve luck (good and bad), the purpose of statistical analysis is to gather enough data so that the luck, which should be random, is filtered out from the "signal".

"I've run over stuff at high speeds in other cars, and I"m here to tell you that they didn't flip over, puncture, or burst into flames...."

It really depends on what kind stuff the car runs over and how fast the car is moving, and also how well the driver is handling the situation when it happens. I've almost run over ladders and mattresses, but never trailer hitch. Maybe I'm lucky.

No it isn't. I've driven over a large rock cast off from an RV going 80mph in a little (2000lbs) sportscar made in the 1960's. There was a large bump, grinding sound and damage to the underfloor (you could poke your finger through).

But in no way was it a major accident.

Please read my separate post about a very similar accident a decade ago that killed 6 children. I don't think anyone can make blanket statements about what is and isn't a serious accident.

A minor accident? Are you sure about that? Running over a trailer hitch at 70 miles an hour is no minor accident. It could be so munch uglier. The car could flips over, tires pop and loss control. In fact, the car continued to operate normally for another minute and stopped safely is quite amazing in that situation in my opinion.

In this case the driver stated that the cabin was undamaged. He was able to remove his papers from the glove box. Smoke inhalation may have been dangerous, but in this case the fire didn't seem to a safety issue to the driver if he had been able to leave the vehicle.

I guess, it would happen the same thing that happened to the papers and pen at the gloves compartment.

This isn't a good argument seeing as how papers and pens can't die from asphyxiation.

I don't see how that is a minor accident. He said he felt the car fly up into the air.

A major accident for me is one in which cars drive oncoming to a semi trailer, hit a power pole at very high speed or multi car pileups. Something where no matter what type of car it is there is likely to be mass casualties and media coverage.

Driving over something at highway cruising speed doesn't seem all that major to me.

Driving over something at highway cruising speed doesn't seem all that major to me.

It's major enough that, as other commenters have noted, it would rip apart the undercarriage of any smaller car, and possibly enter the cabin through the floor.

This is wholly speculative. Floor pans are typically flexible steel. They are not puncture prone. As they are capable of deformatin (absorbs energy). The TSLA has 6mm armor plate, which is stiff and while strong is not tough. It is thus probably not good at disapating energy, so it takes the full brunt. You don't build a bumper out of 6mm armour plate. That being said, maybe TSLA just needs a V-hull to dissipate the force (by deflection, rather than absorbtion). That's how they made the military humvee more impact (blast) resistant, while keeping its armour in tact.

>maybe TSLA just needs a V-hull to dissipate the force (by deflection, rather than absorbtion).

Vhulls are only useful for upwards blasts...short of IEDs showing up in California the concept is useless for a sportcar. Also...humvee doesn't have a vhull - prototypes aside. You're likely thinking of either the MRAP or some buffalo derivative.

Obviously not a "deep v" shape nor something blast resistant, perhaps but some form of deflective geometry might be helpful under the car. It does seem the force ultimately is going ^up^ as opposed to >in< . . .

It might help in a couple of fringe cases like debris hits. You'd murder the aerodynamics though.

Pretty much every other area will suffer too - handling, aesthetics, fuel efficiency, manufacturing complexity, battery replacement etc.

Its just a horrible trade-off.

It would absolutely rip the undercarriage of a small or large car.

But so what ? It would never start a fire.

Never say never. Those things are random. It 100% depends where the object hits. Being the Tesla or any other car. There IS a fuel line. There ARE electronic wires. in any car. And all going underneath the car.

I've personally seen more than 1 regular gasoline car catch fire for various reasons.


Similar kind of accident happened in India, only difference is the passengers there did not get enough time as escape from fires. Here is the link http://www.thehindu.com/news/cities/Hyderabad/45-killed-as-b...

Not real that similar....that bus took out a guardrail.

My point is fire spreads fast and time to escape or no warning signs in other vehicles, where as Tesla has it designed well.

Yes, my petrol engine wakes me up with coffee whenever I'm unconscious.

They should make a C.O.R.A. plugin for the Tesla S onboard electronics. Heck, they should make a K.I.T.T.!

There was a more detailed account floating around. Essentially, each battery cell is in its own fire-resistant compartment. The fire crew that responded to the call used procedures that are standard for other electric cars, but are not proper for a Tesla. Essentially, they "punched through" these compartments in an attempt to extinguish the fire, which actually exacerbated the problem by allowing the fire to spread to other battery cells.

When the fire department arrived, they observed standard procedure, which was to gain access to the source of the fire by puncturing holes in the top of the battery's protective metal plate and applying water. For the Model S lithium-ion battery, it was correct to apply water (vs. dry chemical extinguisher), but not to puncture the metal firewall, as the newly created holes allowed the flames to then vent upwards into the front trunk section of the Model S. Nonetheless, a combination of water followed by dry chemical extinguisher quickly brought the fire to an end.


I saw a Nissan in fire a month or so ago outside of Philly. Car fires happen regardless of engine technology.

It seems putting the onus on every fire department to learn new tesla procedures may be the wrong approach.

Why do you say that? The fire department is a service hired by us to extinguish our stuff. Didn't they learn how to put out gasoline engine car fires 100 years ago? They should have refused because they only knew how to put out horse fires?

In those 100 years, we've learned a thing or two. Which is why we have fire codes and whatnot. We reduce the variation in the number of expected scenarios.

I'll also note that in general, it is easier to learn something different when it's not similar to what you already know. Horse fire vs car fire is easy. Car type A vs car type B vs car type C is more difficult.

Don't joke, horse fires are a very real problem.

Tesla is actually _really_ good at distributing training materials to emergency responders. All vehicle manufacturers do it some some extent, but Tesla sets a very high bar.

As an example, check out this training video they put together: http://www.youtube.com/watch?v=ntK3rvVl2Qw (skip to 26:45 for the fun part ;)

This video in particular is about extrication, not firefighting, but you get the idea...

We train frequently to deal with the new challenges presented by electric vehicles.

Holy hell, that thing cuts through the car's pillar as if it were made of paper!

apparently 2 of 3 fire departments did it wrong, so...

And the 3rd one was in Mexico . . . hmmm

Fire departments had to get special training for the Prius (and other hybrid cars) because of the power cabling going from the rear battery to the front. It's not that unusual.

I think you have it wrong way round - all other electric cars should have batteries split into fireproof compartments

I've never seen a car fire until last week, when I passed by this: http://i.imgur.com/k0xylRe.png. I was really surprised at how large the fire was (it looked worse in person)

Same problem in the WA fire. You cannot extinguish a LiIon fire without (it seems) cutting the car open. In Wa, they put out the fire and it restarted. They had to flip it over and cut holes in it (presumably, through the floor)

see > https://news.ycombinator.com/item?id=6498232

The thing to remember with a battery fire is that it's not enough to just put out the flaming electrolyte, when the battery is shorted out you don't just lose all that energy, it gets converted into heat which is what ignites the electrolyte. The battery in a Model S is made up of 11 modules so when one of these modules is shorted you don't have the entire capacity of the battery being turned into heat but depending on if an entire module was shorted out or maybe two adjoining modules that's still a lot of energy and once the whole thing starts burning you might compromise the other batteries in the system.

Tesla engineered firewalls in between the battery modules but it's not like it's fireproof forever, it just means the heat has to transfer through conduction and gives the driver additional time before the vehicle goes up or before the fire department gets there. What might be a good improvement is a 2.5 inch fire hydrant fitting under the back of the car that the fire department could connect to to flood the battery compartment with water and just hook the car up to a hydrant for 30 minutes to make sure the fire stays out.

> Car fires happen regardless of engine technology.

Of course. But nobody has statistics about the rate of fires per accident type.

And remember that this isn't necessarily a reflection on the engine technology but more so the design of the car. Given that accidents often result in debris and also that people drive over stupid things maybe batteries shouldn't be under the cabin floor.


Its pretty rare for car fires to happen on public highways. Most car fires are incidents with a vehicle at rest, or deliberately set. Something like 2% of accidents involve petrol based fuels systems. Many more are cuased by other issues.

Telsla (as an example) previously recalled ~40% of its Roadster models in 2010 for fire-hazard related faulty wiring. That kind of thing would be attributable to any type of car.

The high incidence of Model S LiIon fires in the field, is much more problematic and worrisome, IMHO. Like you say, it may just be a design flaw on the location of the battery pack. Aft of the front axle is a debri-zone, wheras aft of the rear axle would be more protected from debris (but subject to crumple zone impacts compromizing the integrity of the pack directly or through shrapnel)

The problem with tesla is the active suspension dropping you to a couple inches off the road on the highway. Debris that you would normally pass over shoots into the under carraige..

This is a really helpful and pertinent insight IMO.

A few solutions:

Turn off dynamic highway suspension feature

Use RADAR to detect debris and cause suspension to lift to avoid as much damage (could cause vehicle instability)

>To avoid instability, slowly lift suspension, slow down lift towards apex, lift front suspension first and then rear suspension, and/or slow the vehicle slightly before raising suspension. (Could use rear-facing RADAR to inform decision process about slowing down.)

Of course, with more automation and sensors, the car could possibly change lanes autonomously to avoid road debris all together. (and message the highway patrol and other automated or semi-automated vehicles of the location of the debris as warning)

One last solution I can think of is to somehow either strengthen the undercarriage or add deflectors or crumple zones of some kind.

Personally, I'm thinking that better accident avoidance automation will be the most reliable...including raising the clearance. And making the batteries more puncture-resistant would be great.

On a tangent, would it reduce the likelihood of combustion of the batteries if each compartment could recognize a puncture event and quit drawing power from the batteries in that compartment?

I'm waiting for the day that fully autonomous vehicles are legal and affordable. In the meantime, semi-automous vehicles that "... message the highway patrol and other automated or semi-automated vehicles of the location of the debris as warning ..." is an awesome use of a sub-set of the autonomous technology, and a great way to demonstrate such vehicles' capabilities!

Perhaps they could use a radar scanner at the front to look for oversized objects, and react by pumping the suspension up to maximum height?

Would probably cause instability and loss of traction (at apex) right before a defensive swerve.

By apex I presume you're saying that as the suspension finishes lifting the car, the act of effectively "throwing" the car up into the air will momentarily lower the apparent sprung weight. Makes sense.

If the suspension can act very suddenly, it could execute only when collision is unavoidably imminent.

Does anyone know what the size of this hitch was? What I'm picturing isn't all that large.

mercedes has something sorta similar... it adjust the suspension to adjust for pot holes.

But nobody has statistics about the rate of fires per accident type.

What planet are you on?



Tesla has no engine. Oh, don't follow a truck. http://www.teslamotors.com/models/features#/safety

The reason this was posted by Tesla is so that they control the message. The mainstream media would love to take an incident like this and completely turn in it around to make a news story out of it. But because Tesla has already written about the incident, it makes trying to spin it in a negative light much harder, as Tesla will not need to take a defensive position.

if you crash a regular car and it catches fire do you feel its unsafe? in this case whatever the driver hit must have been pretty big and went through the aluminum (the bottom of the tesla is without any doubt much stronger than a regular car due to the battery compartment). It's likely to have caused a fuel leak and electric fire in any other car as well, except the car wouldn't tell you that and everything would have burnt (if you have never seen a car fire, firemen can't stop the fire from consuming the car once the fire reaches the fuel tank)

the point is that the car knew it was damaged and self diagnosed properly, and that electric circuits are all properly isolated and monitored. Pretty cool if you ask me.

It's not unusual. Many car ads show a crash test.

Trying to get in on some of that TSLA action, eh?

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