Most 'park outside' fire recalls are petrol cars

UL's Fire Safety Research Institute spent nearly $3 million over three years burning six EVs and three fuel-powered cars. The finding: similar fire growth rates, similar peak fire sizes, similar durations. Meanwhile the urgent "park outside" orders keep landing on combustion vehicles.

Nine burned cars, one boring result

The UL Fire Safety Research Institute burned six electric vehicles and three fuel-powered vehicles across three phases of testing, over nearly three years, at a cost of nearly $3 million. This is not a blog post. It is the most expensive controlled comparison of vehicle fire behaviour anyone has published.

The result, per the International Fire and Safety Journal: EV and ICE vehicle fires "demonstrated similar fire growth rates, peak fire sizes and durations during free-burn experiments."

The framing FSRI landed on, reported by Firehouse, is the sentence that should have been the headline everywhere: EV fires are still vehicle fires, with unique challenges around battery access and vehicle construction.

Still vehicle fires. Not a new category of apocalypse. A car burning is a car burning, and cars have been burning for a hundred and twenty years.

The recall list nobody reads

Here is the part that is genuinely under-reported, and CleanTechnica put it in a headline in August: most "park outside" warnings involve vehicles with internal combustion engines.

Not batteries. Seat motors. Power steering wiring. Starter relays. "Always on" control modules that stay energised long after you've parked and walked away.

Go look at the current Jeep Wrangler problem list, compiled from NHTSA data by Pickup Truck +SUV Talk. Recall 26V-363 covers 2021–2025 vehicles whose electric hydraulic power steering pump wiring may overheat and cause a fire — even while the vehicle is parked. That one carries an "Urgent, Park Outside" warning. Stellantis's June action on Wrangler and Gladiator traced fire risk to a power steering pump connection.

The headline on that Wrangler piece, by the way, is "Fuel Pump Failures, Death Wobble, Engine Fires." Nobody made a viral compilation video about it.

One widely circulated recall listicle asserts that fuel-system leaks — rotted fuel lines dropping petrol or diesel onto hot engine parts — are the single most common cause of vehicle fires. We can't verify that from the primary sources we have, so treat it as a claim, not a fact. But note what the claim is about, and note that it never generates a news cycle.

And then there is the failure mode that has no electric equivalent at all. In Bellevue, Washington, police investigated a gasoline theft that sparked a car fire at an apartment complex. In Osceola, Arkansas, a wreck led to a gas station fire that firefighters didn't fully clear until 2:30 p.m. Nobody siphons a battery.

The rate numbers, and why we won't oversell them

The commonly cited figures: EV Infrastructure News puts EV fire risk at 0.0012% against 0.1% for ICE vehicles. SolarCalcPro says EVs catch fire 60 times less frequently than petrol cars.

Those two numbers imply different multiples. That discrepancy is itself the story.

Because here is what Adam Barowy, principal research engineer at the Fire Safety Research Institute, actually said: "Right now there's pretty limited data on whether they are more likely to catch fire — at least for North American cars." Regulators typically have not collected vehicle fire data by powertrain. Starting this year, fire agencies nationwide switched to a new federal reporting system that records powertrain. Before that: fog.

So the honest version. Norway, where plug-in cars are now more than 97% of the new car market, found petrol and diesel vehicles had a fire incidence rate four times higher than EVs from 2016 to 2021. Four times. Not sixty. That's real data from a real fleet.

And NIST Technical Note 2365, published March 2026 and summarised here, used a capture–recapture method to estimate 5,718 battery-electric and plug-in hybrid vehicle fires in the United States since 2011 — with a 95% confidence interval running from 2,866 to 10,846, and excluding Florida entirely because its data behaved strangely. A range with a factor of nearly four in it. That is what "we barely measure this" looks like written in statistics.

As the electric fleet grows, the absolute number of EV fires will rise even if the per-vehicle risk is flat or falling. Anyone who reports the raw count as a trend is either innumerate or hoping you are.

Where the evidence cuts against us

It does, and we're not going to skip it.

EV fires are substantially harder to extinguish and persist considerably longer than traditional car fires. That's not a talking point; it's in the same statistical analysis that gives the 0.0012% figure. Fire and rescue services need specialised training for them. Suppression can take thousands of gallons of water to cool cells and prevent reignition of thermal runaway. The disparity is significant enough that firefighting technique is still evolving.

Batteries also get recalled for fire risk. On the same Wrangler list: 25V-741 and 24V-720, high-voltage battery packs in certain 4xe models that may fail and create a fire risk, with the recall expanding or replacing previous software remedies. LG Energy Solution has faced strict scrutiny in the United States after EV battery-fire recalls. That happened.

The defensible claim is narrow and it is enough: EVs appear to start fires less often, and when they do the fire behaves comparably in size and duration but is harder to put out and needs different tactics. FSRI spent three million dollars to establish roughly that. The 2026 edition of NFPA 70 was updated to keep pace with high-voltage systems and charging infrastructure. This is what a maturing technology looks like.

What the coverage does instead

A Wrangler power steering pump that can ignite in your driveway with the ignition off, across model years 2021 to 2025, gets a recall number. A single EV burning in a car park gets a news cycle and a decade of reposts.

One of those is a fleet-wide defect with an urgent instruction to stop parking indoors. The other is a video.

Six EVs and three petrol cars, three years, three million dollars: similar growth, similar peak, similar duration. The difference isn't how they burn. It's which fires you're shown.

Straight answers

Do electric cars catch fire more often than petrol cars?

No — the available data points the other way, though it is thin. Norway, where plug-in cars are over 97% of new sales, recorded petrol and diesel fire incidence four times higher than EVs from 2016 to 2021. Commonly cited global figures put EV fire risk at 0.0012% versus 0.1% for combustion vehicles, or roughly 60 times less frequent. But FSRI's Adam Barowy is blunt that North American data is limited: US fire agencies only began recording vehicle powertrain in a new federal reporting system this year.

Are EV battery fires harder to put out than petrol car fires?

Yes. EV fires are substantially harder to extinguish and persist considerably longer than traditional car fires, and suppression can require thousands of gallons of water to cool cells and prevent reignition from thermal runaway. Fire services require specialised training and EV-specific protocols. That is the genuine downside — it coexists with EVs starting fires less often.

What did the UL Fire Safety Research Institute EV fire study find?

FSRI burned six electric vehicles and three fuel-powered vehicles across three phases over nearly three years, at a cost of nearly $3 million. In free-burn experiments, EV and ICE vehicle fires showed similar fire growth rates, similar peak fire sizes and similar durations. FSRI's summary: EV fires are still vehicle fires, with unique challenges around battery access and vehicle construction.

How many EV fires have there actually been in the United States?

NIST Technical Note 2365, published in March 2026, estimated 5,718 battery-electric and plug-in hybrid vehicle fires in the US since 2011 — with a 95% confidence interval of 2,866 to 10,846, and with Florida excluded because its data behaved unusually. The width of that range is a measure of how poorly this has been tracked.

Are 'park outside' fire recalls mostly for electric vehicles?

No. Most "park outside" warnings involve vehicles with internal combustion engines, and the causes are usually seat motors, power steering wiring, starter relays and always-on control modules. NHTSA recall 26V-363 covers 2021–2025 Jeep vehicles whose electric hydraulic power steering pump wiring may overheat and cause a fire even while parked; it carries an "Urgent, Park Outside" warning. Battery recalls do exist too — 25V-741 and 24V-720 cover high-voltage packs in certain Wrangler 4xe models.