Highway Vehicle Fires – A Primer On What “Is Normal” And An EV Battery Placement Walkthrough

NOV 11 2013 BY MARC LEE 23

We Know Highway Fires Happen A Lot - But Lets Put Some Context On The Word "A Lot"

We Know Highway Fires Happen A Lot – But Lets Put Some Context On The Word “A Lot”

Much was made over Tesla’s third vehicle fire in 5-6 weeks.  Right leaning media attacked it as sign that EVs are not ready for prime time, and plug-in lovers pushed back that electrics are inherently safer than gasoline fueled vehicles.

So what do we really know?

Systematic record keeping for “highway vehicle fires” (HVF) first began in 1980.  Before that time data was not systematically collected on a national basis.  It is generally thought that the number of deaths from vehicle fires was substantially higher in the 60s and 70s, but surveys and studies from that time come up with widely varying numbers.

Cutaway Of Upcoming BMW i3 Showing Battery Housing

Cutaway Of Upcoming BMW i3 Showing Battery Housing

One such survey in 1973 of government agencies and university research found estimates of  vehicle related fire deaths ranged from 207 to 3500 annually.  Part of the reason for this large discrepancy is that it actually can be difficult to determine whether a person died from the initial accident and then was burned in the ensuing fire or rather the fire itself was the cause of death.  During that time period such a determination was not always or even often made.

Stories from this era of people being burned alive are legion.  Trapped by unyielding seat belts or otherwise pinned in by a deformed post wreck vehicle, horrified onlookers unable to do nothing but witness the gruesome demise.  My own high school Driver Safety teacher recounted a horrific scene of a friend being burned alive because his seat belt would not unbuckle.

Still Waiting on that Mushroom Cloud

One Seriously Destroyed Volt -Still Waiting on that Mushroom Cloud

Gasoline has not always been the number one cause of HVF deaths if  another report is accurate.  It suggests that the primary threat from vehicle fires has “changed from ignition of a small quantity of combustible interior materials by a lit cigarette, in 1960, to ignition of a large quantity of combustible interior and exterior materials by an impact-induced fire.”

Despite the lack of clear data, the need for improvement was recognized and resulted in enactment of  Federal Motor Vehicle Safety Standards (FMVSS) 301 for fuel system integrity in 1967 and and FMVSS 302 in 1972 for flammability of interior materials in passenger cars.

However, in the 1970s, manufacturers were able to get exemptions in certain cases, famously the Ford Pinto was able to get an exemption.  Indeed it was most likely the Pinto case and others like it that lead to the most drastic improvement in motor vehicle fire safety because the calculus of cost to correct unsafe design versus potential legal liability changed dramatically.

The Cadillac ELR And Chevrolet Volt Have The Same Protections For Their 16.5 kWh Battery

The Cadillac ELR And Chevrolet Volt Have The Same Protections For Their 16.5 kWh Battery

In the era of systematic record keeping HVFs appear to have peaked in 1988 with 459,000 fires resulting in 900 deaths and 2750 injuries.  Since 1988 there has been a clear and steady trend of less fires and deaths.  2011 saw just 187,500 fires resulting in 270 deaths and 1020 injuries.

Using the 2011 HVF numbers we can calculate that given a current fleet of plug-ins numbering about 150,000 we would expect to see about 117 plug-in fires per year.  Using 1988 numbers would give 287 plug-in fires.  We have seen three.  Discount the Volt fire because it was two weeks after the fact and was the result of crash testing and also because they failed to discharge the damaged pack, which I would liken to leaving a damaged gas tank full of gas, something they NEVER do.

Likewise discount the one known charger fire because HVFs at gas stations are tallied seperately, and there about 5000 of those each year.

The big caveat to this calculation is that fires tend to occur in older cars when hoses and cables are brittle and frayed with time and wear.  The plug-in fleet is still very young which means the above comparison needs a very big asterisk.

There is no doubt there will be refinements to the structures and systems used to protect batteries.  Gasoline based systems have seen continual refinement for a century but fire based death and injury has not been eliminated.

Tesla Model S Battery Placement Is Low On The Car For A Lot Of Reasons

Tesla Model S Battery Placement Is Low On The Car For A Lot Of Reasons

When you look at the size of the Tesla battery and how low it sits, it is hard to imagine that anything thing that gets popped up into the under belly would not be a direct hit on the battery.  The plate  protecting the battery is substantial compared to typical unibody construction, but obviously it is not going to stop a chunk of metal that is levered up when a wheel rides over it.

It may be that the area between and around the wheels is a common strike area for metal objects that are run over.  The same would likely be true for the engine compart for objects that are struck by the front of the car and levered up.

Nissan LEAF Battery Placement (via expertreviews..co.uk)

Nissan LEAF Battery Placement (via expertreviews..co.uk)

Indeed observing a cut away of the Nissan Leaf you can see that they positioned the battery well away from the front wheels and “engine compartment.”

GM is also looking very smart to have positioned their battery in the center of the vehicle and to have it NOT be the lowest point of the undercarriage.

Ford has the battery on Focus Electric very low to the ground as well, but in their case the battery is positioned more towards the leading edge of the rear wheels.  Presumably a levered object would strike behind the battery if driven over by a rear wheel.

What this all means is that in all likelihood given the knowledge available today, the Tesla Model S is indeed more susceptible to fire than other plug-ins on the road, but compared to internal combustion cars it is much safer – at least according to early data point.

Indeed the fact that no other major model of electric vehicle has ever caught fire on any road in the US means that the Tesla may actually be finding itself penalized from that “pent-up” demand from those anticipating electric vehicle fires.   After all, they are so uncommon, why wouldn’t an EV fire be big news?

Categories: Crashed EVs, Tesla

Tags: , , , ,

Leave a Reply

23 Comments on "Highway Vehicle Fires – A Primer On What “Is Normal” And An EV Battery Placement Walkthrough"

newest oldest most voted

I wonder if the national vehicle-fire database tracks the age of the vehicles involved so that an apples-apples comparison with combustion cars of similar age can be made?

Yes, an apple to apples comparison is needed. Tesla has been disingenuously comparing low mileage 2013 models with the whole population of ICE vehicles without regard for age.

How many 2013 Mercedes Benz have caught on fire after hitting road debris ?

How many 2013 BMWs have caught fire after hitting road debris?

Why do the manufacturers this blog mentions choose a completely different placement for batteries than Tesla. What do they know they Tesla doesn’t, or what Tesla ignored ?

Perhaps because their battery packs are 1/3 or in capacity as Tesla’s. Sounds like the job of a good journalist to find and report on the most relevent statistics. The most relevent being cars of a similar age in a similar environment. Focusing solely on fires caused by road debris only is more misleading than using the readily available, albeit overly general, statistics that Tesla and all NEWS organizations are using.

Battery swap may have been one design consideration for Tesla.

Thanks for an excellent overview!

Has anyone ever died in an EV accident? That Volt crash photo is pretty bad, did they live?


Not in North America, there is a few deaths in Asia, but that is it as far as I know. It is a streak we don’t like to emphasis because we are well past any normal deviation now.

“Ford has the battery on Focus Electric very low to the ground as well, but in their case the battery is positioned more towards the leading edge of the rear wheels. Presumably a levered object would strike behind the battery if driven over by a rear wheel.” That’s not quite accurate. The lower battery pack of the Focus electric is narrower than it’s upper battery with close to a foot of space from either rear wheel, so there is nothing getting wedged between the tire and the battery. Also, the lower battery pack is attached to the same frame as the ICE Focus which is no lower than where a gas tank would be. The issue with the Tesla Model S and road debris could be something as simple as their new Active Suspension, which lowers the suspension to what Tesla calls ‘Low Level’, which is 0.79 under standard height. “Active Air Suspension will automatically lower the vehicle for highway driving to improve aerodynamics. Low Level is also accessible from the touchscreen for loading/unloading of passengers. When the vehicle begins driving the clearance adjusts back to Normal height.” It could be that deactivating that option, and allowing the Model S… Read more »

I think people may be looking at the wrong statistic. The statistic that would be interesting would be the number of accidents caused by road hazards with late model cars. Also any deaths attributed to such accidents.

Reno- +1

While I commend you for writing this, it is incomplete, and I have 2 things: 1) You use the term and include “Plug-Ins”, that would be any vehicle that you plug in to charge the battery. This is not about all Plug-ins as many have ICE also. This is about all-electric vehicles (BEVs), ones that do not have ICE. You also failed to include other all-electric vehicles…the Honda Fit EV, Fiat 500e, Rav 4 EV have batteries on the bottom also, while these are not being produced in the numbers that Leaf, Model S are, they still need to be included. And, all PHEVs, those with ICE, should not have been included here, they still are susceptible to damage to something that could cause gas leak and a much more dangerous situation. 2) The Tesla Model S has the battery on the bottom so that it can be swapped by a machine easily…in 90 seconds, actually. This is ingenious, and, perhaps they didn’t think about what kinds of odd occurrences could happen (how many times do these types of accidents happen, where something large can get under there in such a way as to puncture the battery?). Which, (unrelated), considering… Read more »
Benji I agree there is more data that needs to be looked at to make this complete. However, it is not clear to me that the needed data is available. At least not readily available. 🙂 The overarching point was that, in the not so distant past there were a lot of HVFs. How much we really don’t know but I suspect that 3000-3500 deaths due to HVF per year figure is in the ball park. Despite a century to work on the fuel system protection there are still HVFs in the 100k+ range and hundreds of deaths each year. It is almost inconceivable that there would not be some missteps with battery systems in the early going. I chose to include all plug-ins because in theory anything with a battery could sustain this kind of damage, regardless of whether it also happens to contain an ICE. Indeed as the Volt has shown from its Side Pole Impact Crash Test, its battery too can catch on fire under certain circumstances. I did not mention the “Honda Fit EV, Fiat 500e, Rav 4 EV ” because as you say they are intentionally produced in limited numbers. “The Tesla Model S has… Read more »

FYI: the GM PHEVs you mention and picture here (“Volt”, ELR), they have the battery placement where they do because they are not all-electric vehicles and are using production methods that allow them to combine production with full ICE vehicles of those models. (GM invented the term EREV, marketing bunk for a PHEV, and yes, I understand that the drivetrain is electric, however, it is still a PHEV, not an all-electric vehicle.)

I always have to laugh at the anacronym police around here letting us know which words we should use. When someone says they need to get a new lightening port cable for their iPhone, do you correct them and tell them that it really is just an eight pin connector which carries a digital signal, not actual lightening?

Good article that needed to be written. I agree with a previous commenter that the 120K Plug-In number is misleading. Applying that to just Teslas (figure 20K produced in the past year) makes more sense and is logically more defensible. First, the numbers are annual and the MS has been produced for about a year so we have to take an average – 10K for arguments sake. Using a fire rate of 117/150000 we get 7.8 fires so Model S is well under the limit. However, as you point out this analysis doesn’t take into account the effect of automobile age on the rate of fires. Further, I’m not aware of any data on that. It is highly likely that new cars tend to burn at a much low rate than old cars. So, maybe 3 fires is not that far off of reasonable expectation for the Model S base on the ICE derived data. Though, 3 is such a small sample that one would be foolish to make any deductions. One side point. I’ve heard a number of times the suggestion that one or more of the Model S fires was deliberate. This is wishful thinking. If people were… Read more »

There is no doubt the Model S is an extremely safe car. If it is more susceptible to battery strikes from road debris, that is something that can and should be corrected. But greater susceptibility is far from proven at this point.

Perhaps not allowing the air suspension to lower the vehicle, or prehaps some sort of structural “air dam” just in front of the battery to help clear debris to the side, or who knows what other solution. But as one reader observed if other cars are clearing the debris without incident but the Tesla is so low that it results in an impact which renders the car totaled, well that’s not good.

As this link suggests (https://www.aaafoundation.org/25000-crashes-year-due-vehicle-related-road-debris) there are 25,000 crashes/year due to road debris so the numbers are significant.

Yes. Lots of ideas have been kicked around in the forums. Raising the car higher during highway speeds is an easy one for tesla as they can just push a firmware update. Don’t know if that is sufficient. I doubt a dam is worth a dam (couldn’t resist) – the kind of debris that causes the damage is pretty hard and will probably just slice through the dam on it’s way to the battery pan. Maybe some sort of flak-jacket like armoring could minimize piercing. Probably a combo of things will mitigate the problem.

It’s kind of odd though – I’ve been driving a lot for 40+ years and have never seen the kind of debris that has caused these fires. Not saying it doesn’t exist but in more than half a million miles of not seeing one, I’d say it’s relatively rare. Rarer still is where you don’t see it until it’s too late.

I think the Tesla battery fires should alarm the insurance industry most, as minor road debris might have a high incidence of fire which *may* equal extensive (expensive) repair costs.
This is a hypothesis, rather than a conclusion on my part.
I really really hope I don’t hear of another fire for at least a few years!

I doubt the insurance industry cares much at this point because their biggest issue is injury and death. By far, that this their biggest worry and the Model S has a stellar record in that department.

I think an important distinction should also be noted by the fact that gasoline is combustible in a vastly different way from a battery. Gas just doesn’t smoulder or otherwise have an in-between state between “burning” and “not burning”, while batteries have an “overheating” and “smouldering” and “broken, but not burning yet” state. All of which can be detected by a computer. Leaking gasoline can only be detected by your nose or otherwise the sudden need to escape the flames.

Statistics like the ones being thrown around miss a key issue in making conclusions from numbers: systematics. Some systematic factors have already been mentioned, e.g. late model cars vs. older ones. I suspect that EV drivers and their habits are another systematic: the average Nissan Leaf driver probably drives less aggressively than the average ICE driver, and probably less so than the average Tesla driver as well (wasn’t the Mexico fire after some aggressive driving?). Leaf owners probably spend a smaller fraction of their driving time at high speeds than ICE and Tesla drivers too.

The existing statistics are probably too low to truly sort out systematics like these, but may contribute as much as battery placement/protection to the 3 vs. 0 for Tesla fires vs. other EVs.

Figuring out these systematic factors is the job of actuaries hired by the insurance companies.