Explosions, slashings, crush tests and cars set on fire. Welcome to the National Research Council’s battery abuse lab, our fourth instalment in Electric Autonomy’s Behind the Battery series

Prime Minister Justin Trudeau’s plane had barely cleared the Ottawa International Airport runway when some unusual activity on the ground began.

Gathered around a medium-sized pop-up tent (the variety used as snow covers for equipment during winter) a group of scientist and engineers had been busy moving equipment for an hour, remembers Dean MacNeil, team leader at the National Research Council (NRC) Canada.

Then, soon after the PM’s plane taxied away, everyone went quiet and alert.

For a few moments there was no obvious explanation for the group’s sudden pause — or the three firetrucks standing by.

Then a few wisps of grey smoke began to curl out from under the tent flaps. A few minutes later there was a colossal bang and the fabric tent was engulfed in flames that immediately melted everything but the poles and the scorched metal hull of the vehicle sitting inside.

Firefighters moved in and began to hose down the scene. The onlookers talked amongst themselves and wandered away from the area.

What the PM’s security team (or even Trudeau himself) peering out their plane window might have thought about the fiery scene below them remains a mystery.

But their first guess may not have been that it was a real-life simulation carried out by publicly funded researchers to find out how an EV might behave when it is set on fire.

“This was one of our first forays into very large-scale abuse testing,” says MacNeil, by way of explaining the experiment.

“There was a concern about what would happen if a [EV] battery got exposed to fire.”

NRC Ottawa campus

The graduation to setting EVs on fire grew out of another world-leading area of research in MacNeil’s lab: thermal runaway.

“You have to give the occupant sitting in the vehicle five minutes [to get out] from the detection of thermal runaway,” he explains. “Those are the things we were looking at.”

The good news is that in MacNeil’s EV fire exposure tests (seven to date, in total) there has always been more than five minutes for an occupant to escape. And the fire testing has, says MacNeil, proved that no fire exposure tests are needed for the regulation of EVs. (Combustion vehicles don’t have them.)

But, on the flip side, says MacNeil, “when you abuse batteries, you never really know what’s going to happen.”

So, there is a need for more tests that focus on thermal runaway that happens within the battery.

That is why, across town from the airport and extinguished EVs, MacNeil and his team return to their lab at the NRC’s Montreal Road campus: to continue experimenting with batteries and battery cells.

The setting is a stark contrast to the busy airport. Here, nothing and everything is happening at the same time — and, despite funding it with tax dollars for nearly a century, not many Canadians know it exists.

The complex of roughly 73 buildings spanning 130 acres is the site of many significant Canadian research developments: the explosive RDX; a code breaking and cryptology team called the Examination Unit; and twin “vacuum sphere” wind tunnels for supersonic jet testing, including, once upon a time, the Avro Arrow.

Yet the entire property, with the exception of the security desk, looks almost abandoned on the surface.

But deep in the heart of the campus is building M-48. This is where MacNeil and his team conduct battery stress tests in a lab unlike most others in North America.

“Are we unique? Yes, we are. There’s not many people that can do this in Canada,” says MacNeil. “This is something that we’ve built up since the ’90s. The capability all came from testing for the defence industry. Now, we use it for other analysis.”

The NRC battery abuse lab

What does it look like when you puncture a battery with a nail gun? Crush it, light it on fire, put it in water, freeze it, broil it, slash it, short circuit it or overheat it from the inside out?

Well, if the blackened char marks and divots three feet up the lab’s concrete walls, the soot-clogged industrial vents and the heavy fireproof coverings indicate anything, it is that the NRC battery abuse testing produces dramatic results.

“We are abusing them. This isn’t a real-world scenario. These are extreme scenarios and they’re done for performance reasons,” MacNeil emphasizes. “We’ve learned a lot from everything that we’ve done so that we minimize potential to make mistakes.”

The NRC battery abuse lab has tested cells in big-name OEM vehicles at the behest of Transport Canada and, sometimes, by private companies. They’ve also worked with smaller Canadian players to make sure their battery cell technologies are striking the right balance between cutting edge and cautious.

“I can’t emphasize it enough: safety is an aspect, but it’s not everything. You have to hit other performance metrics. And, so, it’s hitting those other performance metrics, but keeping safety adequate,” says MacNeil.

The NRC also consults with OEMs to equip them with the tools to run battery abuse tests in-house. For obvious reasons, explains MacNeil, “They’d much rather do that test themselves.”

Avoiding thermal runaway

MacNeil says all batteries are measured against the same core standards: cost, performance, durability and safety.

“Thermal management is a huge parameter to get those criteria right. You need high performance and you must maintain ambient temperatures to hit durability metrics,” says MacNeil.

“As you start tearing apart all these battery packs, what you see is that everybody is doing it differently. People come to us with different designs and we look at pushing the limits on those designs.”

In terms of safety, the thing to avoid, says MacNeil, is thermal runaway in an individual battery cell.

Thermal runaway is lab-speak for a heat generating reaction within the battery cell that does not (or cannot) dissipate that heat. “It results in a very rapid temperature rise and the potential of causing a chain reaction to [adjacent] cells,” explains MacNeil in a follow up email.

The result is that when the internal temperature of the battery cell hits around 180 °C “self sustaining exothermic reactions will begin and if they can not cool down, a cell thermal runaway is likely,” says MacNeil.

Without a proper cooling or containment strategy, battery overheating can lead to problems. These include faster degradation, short circuits, venting, fires and, sometimes, even an explosion.

In one of the NRC tests MacNeil and his team held, so much pressure was generated from the battery “venting” due to the heat inside it that it (literally) blew the doors of the abuse chamber open.

“We actually never latch them,” says MacNeil. “Even though [the room is] ventilated. Sometimes it’s too much pressure.”

But these types of situations are rare, assures MacNeil. They have a low probability without outside human interference to excessively stress the battery cell.

And the reason these highly targeted battery cell and pack stressors likely won’t replicate outside the lab tests is because in EVs the packs are put into hard casings designed to protect them from being compromised by external forces.

This is an essential safety layer for consumers and the reason why MacNeil and his team are expanding their facility to include a room where they can test a bigger battery configuration — say a fully assembled pack or an entire car.

” There’s no simple design for battery abuse testing, we test designs for a wide array of applications,” says MacNeil.

“Many times real world events drive our abuse test designs, but ultimately it is our raison d’être to ensure the fundamental science and engineering of the pack design is safe and, if not, to propose potential solutions.”

Burn baby, burn

MacNeil anticipates there will continue to be much more work done on thermal runaway under his watch.

And, depending what research questions need answering, there may be more vehicle tests at the Ottawa Airport down the line. In addition to providing valuable insights for his team, the fires are also helping to train other groups, too.

Firefighters, for example, through MacNeil’s research, are getting valuable experience in managing and extinguishing battery fires safely.

One of the learnings is that it is safer to let a battery fire burn itself out completely rather than extinguish is partway through, as long as there is no danger to adjacent infrastructure. This is because some of the ingredients within a battery cell remain unstable in partially burnt battery packs. This increases the chance of the fire reigniting without much warning.

‘It’s called stranded energy,” says MacNeil. While stranded energy is present in all batteries, a complicating factor is that how it behaves can vary between chemistries.

“If you put some new pixie dust in the electrolyte, pixie dust will be in the fire. So, it’s a scenario you have to watch,” says MacNeil.

The pixie dust comment leads to the all-important question: is there one battery chemistry that’s safer than the rest?

Not exactly, says MacNeil. Each chemistry has pros and cons in terms of safety management. That will likely be true in future as the technology evolves.

But the EVs on the market are, thanks in part to the tests conducted in MacNeil’s lab, very safe.

“You don’t want to over regulate a new technology. You want to make sure your baseline safety metrics [for EVs] are the same as the ones used for internal combustion vehicles,” says MacNeil.

“I do this for a living and I still buy electric.”