Hacker News Reader

Comment by abalone | Hacker News Reader

This is all very good PR.... BUT. The critical safety concern with the Tesla is.......

You guessed it. The high voltage system. It can kill you or a first responder with a touch.

It's not just about batteries catching fire, although that's gotten a lot of press lately. A chief concern during a Tesla accident is the exposure of high voltage components.

It's a big concern for first responders. Read it from the horse's mouth, the Tesla Emergency Response Guide. Page 16 highlights the "no cut" zones of the car. It's practically the entire length of the car.

http://www.teslamotors.com/sites/default/files/downloads/201...

There was a video they put out as well that details the steps first responders need to take to cut someone out of a car. It's HELLA COMPLICATED and in very close proximity to dangerous high voltage components.

And yes, gas cars can explode, etc. I don't mean this as fuel for the gas vs. electric debate (no pun intended). I hope very much the industry can improve electrical safety over time. All I'm saying is, never trust Elon Musk PR to tell you the whole picture.

View on HN

> All I'm saying is, never trust Elon Musk PR to tell you the whole picture.

Never trust anyone's PR to give you the whole picture!

Though as a counterpoint, I've seen cars with the roof cut off to extract people, but never the side doors or the floor cut open. I don't think they'd cut the floor since it's quite thick and has all sorts of wires and pipes anyway.

You've never seen doors cut off? You must not have seen much then.. happens all the time. Google jaws of life.

BTW here is the Tesla first responder training video that shows how to cut off a door(!) and move the dash without electrocuting yourself. http://www.youtube.com/watch?v=ntK3rvVl2Qw

It's a pretty cool video, I expect most people on this thread would enjoy watching it. I sure did.. although I'm not sure I'd want to be stuck bleeding in a Tesla while they figure it out.

Also a little concerned about a short happening during an accident and electrifying passengers.. would like to know more about the risks there.

Firefighter here...

The major thing that concerns me about the Model S is the placement of the DC-DC converter in the front right wheel well. There's a common maneuver called a 'dash lift' where you make a couple relief cuts and then lift the dash using a hydraulic ram. Properly place, those cuts are dangerously close to that high voltage system.

Explanation of relief cut placement: http://advancedextrication.com/2011/12/closer-dash-lift-part...

EDIT: The video abalone posted has a great walk-through on rolling the dash of a Model S. http://www.youtube.com/watch?feature=player_detailpage&v=ntK...

Another issue is the seat belt pretensioner, which would complicated a 'side blitz' door removal (where both doors are removed as a unit), but that's a pretty standard place for the pretensioner, so it's something we're used to working around.

Looks challenging. How does that compare with a Prius?

A Prius is a lot simpler. All the electric components are well away from all the 'usual' cut locations.

Obviously all the normal concerns apply (air bags, pretentioners, etc), but from an extrication perspective, they're pretty straightforward.

The one issue common to all EVs (hybrid or plugin) is that the can silently deliver power to the wheels. With a 'normal' car, you can hear the engine running. With an EV, it can be harder to tell if it's on or not.

One of the first things we do it secure the keys, but with keyless ignition, the key may be trapped in the drivers pocket. That means the big round 'On' button just needs to be bumped by someone for the car to power up. That's why it's critical that we're able to disable the electrical systems of the car, to prevent it from starting up unexpectedly.

> I hope very much the industry can improve electrical safety over time.

With regard to the electrical aspects, isn't the safety record on these cars perfect or close to perfect up until this date? That means it's only going to get worse. That does not mean electrical cars are unsafe, it just means there isn't much of a record yet.

My expectation is that there will over time be a small number of electricity-related accidents involving first responders and passengers and that they will be enough to fuel concern trolls until the end of time. That the electrical problems will be so enormous that they dwarf the massive safety gains from better passenger protection of the sort NHTSA is interested in (and frankly, anyone who is about to get hit by an oncoming SUV is interested in) seems awfully counterintuitive.

Wish I had your crystal ball.

With improvement, I was referring to the complexity of cutting someone out of a Tesla and all the no-cut zones that could electrocute passengers, emergency responders, even guys towing wrecks. You're right, there is no real safety record yet.

> With improvement, I was referring to the complexity of cutting someone out of a Tesla and all the no-cut zones that could electrocute passengers, emergency responders, even guys towing wrecks.

As an electrical engineer with spacecraft experience, I can say that a car like the Model S isn't likely to electrocute anyone. While extracting someone from a wreck, emergency workers use tools like the "jaws of life", strong metal pliers meant to quickly snip a car's body apart.

When a tool like this cuts into an electric car, if it should slice through a power cable, in most cases the tool would short the cable to the car's chassis, thus grounding it and tripping the safety links that are integral to the battery pack. This means the car will become inoperative, but it doesn't mean anyone will get electrocuted.

I think the above will be the outcome in the vast majority of cases. Compare this to the plight of someone sitting in a wrecked gasoline-fueled car.

Hi there. Go ahead and watch this video from Tesla about how to cut someone out of a Tesla. The very first step is to cut the circuit like you say, but there are a few things you overlooked.

Much of the video is concerned with instructing responders how to cut around components that may remain electrified such as the converter and capacitors which, in Tesla's own words, present a "serious hazard of shock" because "they can release up to 400 volts in just an instant."

http://www.youtube.com/watch?v=ntK3rvVl2Qw

Do other car manufactures provide such videos?

I know a paramedic in rural Australia. They get taken to local car dealers to get safety training with electric cars.

As I understand it it's done on a per-model basis, but that wont really work as an increasing number of electric car models end up on the roads. I'm not sure if this is done all over the state/country, or if it's the one local dealer being friendly with the local emergency services.

Nitpicking, I know, but releasing 400 volts means nothing. Current kills, tension does not. It all hinges on the internal resistance of the power source.

That and time. You could get hit with high volts and current, but still be fine if the exposure is short. I have had a few 240v mains shocks on high current fuses, but they were all pretty quick and I had nothing but very slight burns and sore muscles for about ten minutes.

However, there is definitely a risk here, albeit what appears to be a manageable one. Perhaps there could be some use in building volt meters that glow into gloves for emergency responders to help mitigate some of the risk.

Perhaps it would be even better to say, "power transmission kills."

Well, also the resistance of the load.

Voltage means A LOT. A regular 12V car battery is a perfect proof: it has enough current to kill you, but at only 12V it doesn't overcome your body's resistance much under most circumstances.

400V can overcome your body's resistance no problem. And obviously an EV has a crap ton of amperage as well.

Would the tools used not be developed?

I would've assumed the production of specialised tools for electric cars would be a priority once they become popular.

Presumably, avoiding electrocution of the passenger being extracted is also a goal.

I don't understand your response, is this not what the specialised tools would be primarily designed for?

What sort of tools are you envisioning?

maybe insulated ones

Insulating the tools isn't really the issue. There's already a fair amount of insulation between me (the tool operator) and the business end of the tool.

The problem is, we're cutting through and crushing various bits of the car that weren't really designed to be indiscriminately cut into (which is why training is so important).

Both rescuers and victims have been seriously injured and killed when extrication activities turned potential energy stored in the car into kinetic energy. Air bag deployment systems, seat belt pretensioners, even the little hydraulic tubes that hold your trunk open and be extremely dangerous if their energy is released in an uncontrolled way.

So you can understand why we're a little leery of 85 kWh of potential energy sitting under the car...

Here's an example of airbags being deployed when the SRS controller was crushed with a hydraulic spreading tool.

https://www.youtube.com/watch?v=yDSksTztHJk

Two firefighters were injured in that incident (one of them seriously).

85 kWh of potential energy sitting under the car...

A 40 litre petrol tank holds about 380 kWh though.

A 40 liter petrol tank can't discharge 85kWh of potential energy in seconds without ideal conditions (good vaporization or atomization combined with sufficient oxygen)

Neither can a battery. The internal resistance is too high. They can cook and go on fire a bit, but much less than a gasoline fire, but they cannot shock you with their entire capacity in a few seconds.

Yeah, but we've had almost a century of experience dealing with that...

The issues involved are also very different. A tank of gas is actually a very stable thing. Barring a puncture, there's not much you can do to a gas tank to make it do bad things.

Cutting a gas line is no major concern. Shorting a high voltage/high current power line to the frame of the vehicle is putting a lot of trust in the batteries failsafes (that trust is almost certainly justified, but with a track record measured in years, not decades, you'll forgive us for being a little paranoid...).

As far as I can tell, unless people have done something stupid, your worst worry is not the shock from the batteries. The batteries are a variety of potential difference spread over a large area and will go on fire rather then discharge everything at once, which is an issue, but they will not explode with their total chemical potential, not even close. It is any big fat capacitors that you have to worry about. Now they are not storing anywhere near the capacity of the battery, but they can let it out with very low internal resistance and after heavy braking are likely to be fully charged. So it is a current that can be easily discharged, but you have to know it is there.

edit - I said it earlier somewhere else, but EMT gloves with built in glowy multimeters might be a really good plan.

double edit - if you think this is a good plan, feel free to nick it.

Would an independent accelorometer in the battery assembly shut of the power when the car is involved in a collision?

It wouldn't make sense for it to be dependent on a link from the airbag sensors, it would have to be local to the power plant.

Aren't those the same concerns people had when gasoline cars were introduced? It was considered to be too dangerous to use.

Fast forward to today, cars are not exploding. Battery safety issues will be dealt with.

Considering the fact that Tesla batteries are made for lots of really tiny cells, I can't imagine it is impossible to disconnect them from each other in the case of a crash.

I don't know about individually disconnecting each cell, but there is a loop that emergency responders are supposed to manually cut to disconnect the battery. However there are still components that retain power and remain hazardous such as the converter and capacitors. Firefighters need to take great care to cut around them. Plus there's the possibility of the battery compartment being compromised in an accident as well. Tesla's training video covers all this.

Nobody's arguing that Teslas are unsafe, just that the "safest car" claim sidesteps the electrical safety issue. You can't really argue that everything's fine because nothing bad has happened yet, because there are so few Teslas on the road. And it should be noted, a lot of people were hurt along the path to making gasoline cars as safe as they are today. Look at Ralph Nader's work on automobile safety for example.

I actually believe progress demands that we work through electrical car safety. But what irks me is it's just a little premature to be trumpeting "SAFEST CAR EVAR" when you've got a novel set of serious safety issues that the test suite doesn't even try to target.

Does anyone know what the actual voltages are?

Just because it is high relative to the usual 12V of cars does not mean it is terrifically fatal.

An anecdote to give you a picture of things: Electric arc welders operate at around 60V with a very large amount of current. However the electricity is not particularly the dangerous thing about them, it is the 9000F arc you are making and the molten steel. So to demonstrate that we shouldn't freak out about the electric danger and worry more about the heat, my instructor had someone touch the business end with one hand and the ground electrode with the other. Bit of a tingle but certainly didn't stop their heart or anything. Welders are kept at low voltage for this reason.

I assume that it must be somewhat dangerous because of the responder guides warning, but I would want to know the actual voltage before drawing conclusions. If the voltage is 112 or less (or even 220 or less), it should be possible to be pretty safe with some fairly manageable precautions.

These are very high voltages. The Tesla is 375 volts and the Prius is 275 volts.

Technically, that puts it well into the 'low-voltage' classification; generally, 1000 V is the threshold for 'high-voltage'.

> generally, 1000 V is the threshold for 'high-voltage'

DCV? Not an expert, but my understanding was ~1000V for AC, but much lower for DC

The definition depends on the context. When discussing risk of electrical shock, below 50V is usually considered low voltage (and safe).

You can kill yourself with much lower voltages than that, assuming the power supply is capable of delivering the amps.

Wow, I didn't realize they went that high. Then yes, with the current output of a battery pack, I can see why people are worried.

The voltage is not particularly the issue. Petrol cars have a high tension line for the spark plugs that goes to over 30,000 volts.

About that, do cars attempt to shut down engines, fuel intake and electric pathways when detecting a collision ? I remember Liquid Gas vehicles having an emergency outlet to avoid fire/explosions. In Tesla's case that would be cutting the battery off the circuit and maybe putting it into a 'fast drain' mode ..

Fast draining those batteries would likely cause an enormous amount of heat.

This.

Batteries have internal resistance. They're generally rated for a certain level of drain (amperage), which, if exceeded, will go over the thermal limits for the battery.

In the case of a Tesla, you'd risk being fried (current) _and_ roasted (heat) at the same time.

Better simply to keep the energy contained.

You can't just "shut down" electric pathways. Not exactly. There is a manual cut that first responders need to make to sever the high voltage circuit before they can work on the car. However there's still capacitors, converters and the battery itself that remain dangerous especially if they've been compromised.

As for "fast draining" the battery, where would the power drain to? That's exactly the problem.. NOT giving the power a path to go when you're working on the car.

This is not much of an issue with most cars because the electrical systems are low voltage (12V car battery). With Tesla it's enough to cause major injury or death with one touch if you complete the circuit.

Watch the video, it explains all this. http://www.youtube.com/watch?v=ntK3rvVl2Qw

And I guess it's unrealistic to think of a special draining device to be brought on crash sites to exhaust the car battery. All in all, are electric vehicles more 'instable' energy wise when crashed, compared to petroleum based one ? or is it different flavours of similarly sh*tty situations ?

That's a very big question and probably only time will tell. Without getting into it too much, a very basic difference is..

Gas can explode like a bomb but (a) it needs an ignition source and (b) there can be visible signs when the fuel is leaking to let you know to take extra precautions (like spray fire retardant). You can actually get fuel on you and you'll be fine.

High voltage systems don't (usually) explode, but they have lethal amounts of energy stored that can find its way into your body if you complete a circuit. Furthermore in a mangled wreck there aren't necessarily visible signs when a component is electrified. You find out the hard way.

Basically Tesla deserves props for making an extremely strong car. That's what was tested here. But the spin is they're claiming "safest car" while completely side-stepping the high voltage safety issue.

> Gas can explode like a bomb but (a) it needs an ignition source and (b) there can be visible signs when the fuel is leaking to let you know to take extra precautions (like spray fire retardant).

And (c) if it's a diesel car (very common in europe), good luck getting it to explode.

Want to scare your friends? Put out a lit match or cigarette using Diesel fuel.

Diesel fuel requires a lot of heat to combust. Diesel engines don't contain spark plugs, they heat the fuel by compressing the air that the fuel is atomized in.

Technically they only claimed "ACHIEVES BEST SAFETY RATING OF ANY CAR EVER TESTED" which is completely true.

The "unaware of any Model S or Roadster occupant fatalities in any car ever" is also a pretty strong argument for actually being the "safest car" too.

>Basically Tesla deserves props for making an extremely strong car. That's what was tested here. But the spin is they're claiming "safest car" while completely side-stepping the high voltage safety issue.

That's not a side-step. IIHS doesn't test for this, and Tesla is not beholden to do so.

There isn't really a good way to quickly and safely dump all the energy from a battery the size of the Tesela's. Think about it: to drain in an hour, the battery would have to source 65kW continuously, converting all that energy into either heat or light somehow.

85KWh is a lot of energy. Back of the envelope calculations puts that at enough to boil around one metric ton of water from 25C.[1] If you wanted to rapidly drain a Tesla you would have to dump that energy somewhere.

That might actually be plausible for firefighters to do[2], but I doubt the battery itself would withstand being drained anywhere near fast enough for an emergency situation. It would probably find some interesting way to fail if you forced it to discharge that fast (bypassing all the safeguards).

[1] 3.06e8 J / (4.181(J/(g * k)) * 75K) = 976,000g ...I probably got that wrong, it's been a long time.

[2] I mean, their main job is getting rid of large concentrations of heat, right? ;)

Also remember we're talking about an accident scenario. There's no way emergency responders would intentionally run a high voltage circuit like that where there's a risk of short circuit in a physically compromised vehicle.

They're going to be focused on containing the energy not draining it.

It is fun to do back of the envelope calcs though.. You could run 50 space heaters for an hour off that sucker. Insane.

> They're going to be focused on containing the energy not draining it.

In an emergency that involves extracting passengers from an electric car, the most likely outcome is that the battery will get shorted to the car's chassis, at which point the emergency links in the battery will open up, isolating the battery from the rest of the car and its passengers.

As the video explains, even when the battery has been isolated there are other components that remain electrically hazardous such as capacitors and the converter.

Also there is the possibility that the battery itself may be physically compromised in an accident, e.g. it could protrude into the cabin (it sits right under it). Not much you can do to isolate it then.

Caps hold energy, sure, but if you're cutting through one it will self-short, which will be momentarily unpleasant but should not be close enough to anyone to actually injure them. What would be more worrying would be to cut through a circuit without completing a current flow path for the capacitor as that would leave open the possibility of someone accidentally completing the circuit themselves. But even this is not a large risk if you use conductive tools (which will absorb the brief current surge) or wear anything insulating.

The concern with the inverter-converter is mostly due to the fact that it's hooked up the battery. If the battery is already open-circuited then the inverter-converter should just be a dumb block, modulo any residual energy left over in a rectification circuit or something.

> It is fun to do back of the envelope calcs though.. You could run 50 space heaters for an hour off that sucker. Insane.

Wolfram Alpha puts it at 1/5th the energy in a typical lightning bolt... Absolutely nuts.

That's the amount of energy needed to bring nearly 1000 kg of water from 25°C to 100°C, but not enough to begin the phase transition from liquid to gas (i.e. start boiling).

The heat of vaporization of water (i.e. the amount of energy required to convert liquid water at 100°C to vapor at the same temperature) is 2260 J/g. So to completely boil your metric ton of water at 100°C would require an additional 1000 kg * (2260 J/g) = 2.3 GJ, which is 627 kWh!

At 85 kWh, the Tesla's battery contains enough energy to completely boil off 119 kg of water starting at 25°C [0], which is still pretty impressive!

[0] 85 kWh / ((4.181(J/(g * K)) * 75 K) + (2260 J/g))

Ah very true. Knew I was forgetting something.

While that's true, I don't think it's what the parent meant. Colloquially, when we say to boil water we don't mean until the vessel is dry, just until a roiling boil is reached (this is typically how we cook pasta).

If the 1000 kg is heated uniformly, this is not enough energy to bring it to a "roiling boil". There will be no bubbles.

Right. It's a good thing stoves/ranges heat from the bottom!

You're forgetting convection.

And it won't get easier since "we" want battery capacity to increase.

Most cars automatically shut down fuel supply to the engine when a collision is detected. Even if it's just a minor hit the engine can't be started again without clearing the collision warning from the computer.

Thanks, now I wish to know more about the protocols and checks in car controllers.

Shut down fuel supply is code for turn off the fuel pump.

This can be done all mechanically, scroll down on this page to see a deconstructed inertia switch. http://www.therangerstation.com/tech_library/InertiaSwitch.h...

About that, do cars attempt to shut down engines, fuel intake and electric pathways when detecting a collision ?

Some do, and a lot of people don't know about that. It can trip in small accidents too so sometimes people can figure out why their car won't start when it seems perfectly fine. Check your manual to see if you have it and how to reset it, it's good to know.

That said, my grandfather was in an accident where he slid off an icy roadway and flipped his Toyota truck. It didn't trip the cutoff (may not have had one) and the engine continued to run upside down. He had to reach in (after he had crawled out) and turn off the ignition. It surprised him because he had never owned a car that was fuel injected.

Recycling the same astroturfing from the Prius a decade ago. Boring. Try something new, or people remembering last time around are going to LOL not get scared as intended.

While it's true these concerns are probably overblown in the eyes of the public (it's not something they need to worry about). It's absolutely not 'astroturfing' from my perspective (as someone who may need to cut one of these up some day).

I very much appreciate the amount of effort Tesla has put into producing training materials for working with their vehicles.

Good thing regular cars don't contain high voltage components or volatile, flamable fluids that can cause intense fires and even explosions...

Car explosions have been thoroughly busted by Mythbusters. They are as busted as "Enhance!"

Car fires are very real and nasty, though.

High voltage? Do you mean the 12V battery that can't shock you, or the spark ignition that won't do much more than startle you?

I've had numerous sparks jump from distributors and plug wires through my hand. They just make your hand numb/tingly. If the jump went in your left hand and out your right, there might be some cardiovascular risks, but that's nothing compared to the risks of 400V @ god knows how many amps.

Reply on news.ycombinator.com